Proceedings of ICT, SOCIETY AND HUMAN

call for papers

IADIS MULTI CONFERENCE ON COMPUTER SCIENCE AND INFORMATION SYSTEMS

21 - 23 June Algarve, Portugal

Proceedings of ICT, SOCIETY AND HUMAN BEINGS 2009 and WEB BASED COMMUNITIES 2009

Edited by:

Gunilla Bradley Piet Kommers

international association for development of the information society

IADIS INTERNATIONAL CONFERENCE

ICT, SOCIETY AND HUMAN BEINGS 2009 and IADIS INTERNATIONAL CONFERENCE

WEB BASED COMMUNITIES 2009

part of the IADIS MULTI CONFERENCE ON COMPUTER SCIENCE AND INFORMATION SYSTEMS 2009

ii

PROCEEDINGS OF THE

SECTION I


ICT, SOCIETY AND HUMAN BEINGS 2009 SECTION II


WEB BASED COMMUNITIES 2009 part of the IADIS MULTI CONFERENCE ON COMPUTER SCIENCE AND INFORMATION SYSTEMS 2009

Algarve, Portugal JUNE 21 - 23, 2009 Organised by IADIS International Association for Development of the Information Society iii

Copyright 2009 IADIS Press All rights reserved This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, re-use of illustrations, recitation, broadcasting, reproduction on microfilms or in any other way, and storage in data banks. Permission for use must always be obtained from IADIS Press. Please contact [email protected]

Volume Editors: Gunilla Bradley and Piet Kommers Computer Science and Information Systems Series Editors: Piet Kommers, Pedro Isaías and Nian-Shing Chen Associate Editors: Luís Rodrigues and Patrícia Barbosa

ISBN: 978-972-8924-82-9

SUPPORTED BY

iv

SECTION I


ICT, Society and Human Beings 2009


v

vi

TABLE OF CONTENTS

FOREWORD

xi

PROGRAM COMMITTEE

xiii

KEYNOTE LECTURE

xvii

PANEL

xviii

FULL PAPERS BUILDING TRUST ONLINE: DO PRIVACY POLICIES HELP OR HINDER?

3

Margaret Tan

CHALLENGES OF THE DEVELOPMENT OF A KNOWLEDGE ECONOMY AT GRASSROOTS LEVEL: CASE STUDY OF A CULTURAL CENTER IN SENEGAL

11

Anne-Marie Touru and Tuija Tiihonen

EXTENDING A MEETING MODEL AND IDENTIFYING SOCIAL ACTIVITIES FOR APPLICATION SUPPORT

20

Terence Blackburn, Kuan Tan and Paul Swatman

CONTEXTS FOR COUNSELLING - OPPORTUNITIES BY COUNSELLING VIA TEACHERS’ WEB FORUM

28

Jarle Sjøvoll

REFLECTIONS UPON ETHICAL CHALLENGES OF IT-PROFESSIONALS

35

Darek M. Haftor

IT PROFESSIONALS’ PERCEPTIONS ON THE NATURE OF MORAL PROBLEMS

43

Tero Vartiainen

DIGITAL IMMIGRANTS, DIGITAL NATIVES AND CONTINUING PROFESSIONAL DEVELOPMENT (CPD) FOR ACADEMICS: ARE THE STEREOTYPES UNHELPFUL?

51

Andrew Rothwell

INVESTIGATING UNIVERSAL ACCESS FROM A HUMAN DEVELOPMENT PERSPECTIVE Florence Nameere Kivunike, Love Ekenberg, Mats Danielson and F F Tusubira

vii

58

THE HARE AND THE TORTOISE: A COMPARISON OF NORTHERN AND SOUTHERN HEMISPHERE FINDINGS OF THE SITES 2006

67

Seugnet Blignaut and Christo Els

ANTI-SOCIAL NETWORKING SITES: WEB 2.0 AND SOCIAL EXCLUSION

75

Denise Leahy and Ultan Ó Broin

UNDERSTANDING HUMAN STATES CONDITION BASED ON PHYSIOLOGICAL AND PSYCHOLOGICAL INFORMATION

83

Santoso Handri, Shusaku Nomura and Kazuo Nakamura

AN APPLICATION OF ROUGH SET ANALYSIS TO A PSYCHO-PHYSIOLOGICAL STRESS RESEARCH

91

Shusaku Nomura, Santoso Handri and Yasuo Kudo

COLLABORATIVE TEAM OBSERVATION AND SUPPORT

98

Kuan Liung Tan and Paul Swatman

E-GOVERNMENT FOR OLDER USERS: A NEW ZEALAND CASE STUDY

105

Kay Fielden and Pam Malcolm

GPS MOBILE TECHNOLOGY; AN INVESTIGATION REVEALING LONG-TERM IMPLICATIONS TO SOCIETY

113

John Necessary and Fred L. Kitchens

A MODERN FETISH: THE VALUE OF THE MOBILE PHONE IN SOUTH KOREAN YOUTH CULTURE

121

Jukka Jouhki

DESIGN OF A KNOWLEDGE SHARING SYSTEM IN A NON-PROFIT ORGANISATION

129

Kuan Liung Tan, Paul Swatman and Aditya Pethe

E – GOVERNANCE: A FRAMEWORK FOR CITIZENSHIP

139

Krassimira Paskaleva

AN INFORMATION PRODUCTION AND OWNERSHIP PLATFORM FOR BOPERS

147

Ashir Ahmed

SOCIO-TECHNICAL CAPITAL AMONG TWO VIRTUAL COMMUNITIES FOR PEOPLE WITH MULTIPLE SCLEROSIS

155

Maire Heikkinen

MOBILE PHONES AS HIGH IMPACT, LOW COST TOOLS, IN THE FIGHT AGAINST HIV/AIDS WITH MEN THAT HAVE SEX WITH MEN AND MALE SEX WORKERS Christopher S. Walsh

viii

163

SHORT PAPERS EXPLORING TRAINEE TEACHERS INFORMATION AND COMMUNICATIONS TECHNOLOGY (ICT) LITERACY LEVELs : IMPLEMENTATION OF A SMART SCHOOL MODEL

175

Jessnor Elmy Mat-jizat and Elspeth McKay

WHAT ABOUT THE ‘NET GENERATION’? - TRYING TO NUANCE THE STEREOTYPICAL IMAGES

180

Sheila Zimic

BRIDGING THE DIGITAL GAP WITH PUBLIC ACCESS TO ICT: RESULTS FROM A BRAZILIAN SURVEY

184

Esther Menezes, Graziella Cardoso Bonadia and Cristiane Midori Ogushi

MEASURING THE RESISTACE TO CHANGE IN THE WORKLOAD ALLOCATION SYSTEM INTRODUCTION PROCESS

189

Nathalie Hrycej and Kathy Buckner

INDIVIDUAL “ICT LANDSCAPES” AND QUALITY OF LIFE

193

Pernilla Gripenberg

THE ROLE OF INFORMATION SYSTEMS IN INFORMATION MANAGEMENT: EXPLORING THE CHALLENGES IN INFORMATION SHARING

198

Seppo Sirkemaa, Karen Anderson, Karl W. Sandberg, Olof Wahlberg and Erik Borglund

SOME GUIDELINES FOR THE DESIGN OF INFORMATION SYSTEMS FOR COMPLEX DECISION MAKING: PRELIMINARY RESULTS

203

Pascal Vidal, François Lacroux and Darek M. Haftor

3D VIRTUAL WORLD CREATES NEW GENRE OF CYBER SEX

209

D. Yvette Wohn

SPECIALIST KNOWLEDGE MANAGEMENT

213

Mounir Kehal

REFLECTION PAPERS DOVETAILING WITH DIVERSITY: A CHALLENGE FOR HCI

221

Sebastiano Bagnara and Simone Pozzi 225

ROLE OF ELECTRONIC RECRUITMENT Seppo Sirkemaa

TOWARDS IMPROVED CONCEPTS OF APPROPRIATE USABILITY

229

Barrett S. Caldwell 233

THE INTERNET, OUR CYBORG MIND Nuno Nodin

ix

INTERNATIONAL MOBILITY OF WOMEN IN ICT CAREERS

237

Núria Vergés Bosch, Ana María González Ramos and Rachel Palmen

EMPIRICAL STUDY OF ROOT CAUSES OF CHANGE IN IT IMPLEMENTATION PROJECTS

242

Ossa Elhadary

POSTER 249

WEB COMPETENCIES IN SME Klára Antlová, Tomáš Gregorovič, Jindřich Tandler and Lubomír Popelínský

AUTHOR INDEX

x

FOREWORD These proceedings contain the papers of the IADIS International Conference ICT, Society and Human Beings 2009, which was organised by the International Association for Development of the Information Society in Algarve, Portugal, 21 – 23 June, 2009. This conference is part of the Multi Conference on Computer Science and Information Systems 2009, 17 - 23 June 2009, which had a total of 1131 submissions. The effects of ICT on humans comes in focus in the conference but also the interaction between ICT – Individual – Society. Interaction and Effects - both are important. Changes in behaviour, perspectives, values, competencies, human and psychological aspects, feelings are all of interest. Computer Science and IT related disciplines work more and more together with various behavioural sciences: Cognitive effects and changes, Motivational and emotional changes, Organisational and institutional changes, Societal changes, Value changes and new lifestyles, experiences from Child psychology and Development psychology. In general all types of research strategies are encouraged and especially cross disciplinary and multidisciplinary studies. Case studies, broader empirical field studies, theoretical analyses, cross cultural studies, scenarios, ethnographic studies, epistemological analyses. The IADIS ICT, Society and Human Beings conference addresses in detail eight main aspects: Globalization and ICT, Information and Communication Technologies (ICT), Life environment and ICT, Life role and ICT, Effects on humans, Actions for reaching the Good Information Society, Desirable goals and Perspectives. The conference has the intention to provide a contribution to academics and practitioners. So, all are fundamental and applied research are considered relevant. The IADIS ICT, Society and Human Beings 2009 conference received 92 submissions from more than 28 countries. Each submission has been anonymously reviewed by an average of four independent reviewers, to ensure that accepted submissions were of a high standard. Consequently only 21 full papers were approved which means an acceptance rate below 23 %. A few more papers were accepted as short papers, reflection papers and posters. An extended version of the best papers will be published in the IADIS International Journal on WWW/Internet (1645-7641) and IADIS International Journal on Computer Science and Information Systems (ISSN: 1646-3692) and also in other selected journals, including journals from Inderscience.

xi

Besides the presentation of full papers, short papers, reflection papers and posters, the conference also included one keynote presentation from an internationally distinguished researcher. We would therefore like to express our gratitude to Professor Dr. Lorenz M. Hilty, Head of Laboratory, Technology and Society Empa, Swiss Federal Laboratories for Materials Testing and Research, Switzerland, for accepting our invitation as keynote speaker. As we all know, organising a conference requires the effort of many individuals. We would like to thank all members of the Program Committee, for their hard work in reviewing and selecting the papers that appear in the proceedings. This volume has taken shape as a result of the contributions from a number of individuals. We are grateful to all authors who have submitted their papers to enrich the conference proceedings. We wish to thank all members of the organizing committee, delegates, invitees and guests whose contribution and involvement are crucial for the success of the conference. Last but not the least, we hope that everybody will have a good time in Algarve, and we invite all participants for the next year edition of the IADIS International Conference ICT, Society and Human Beings 2010, that will be held in Freiburg, Germany. Gunilla Bradley, Royal Institute of Technology Sweden ICT, Society and Human Beings 2009 Conference Program Chair Piet Kommers, University of Twente, The Netherlands Pedro Isaías, Universidade Aberta (Portuguese Open University), Portugal Nian-Shing Chen, National Sun Yat-sen University, Taiwan MCCSIS 2009 General Conference Co-Chairs Algarve, Portugal June 2009

xii

PROGRAM COMMITTEE ICT, SOCIETY AND HUMAN BEINGS CONFERENCE PROGRAM CHAIR Gunilla Bradley, Royal Institute of Technology, Sweden

MCCSIS GENERAL CONFERENCE CO-CHAIRS Piet Kommers, University of Twente, The Netherlands Pedro Isaías, Universidade Aberta (Portuguese Open University), Portugal Nian-Shing Chen, National Sun Yat-sen University, Taiwan

ICT, SOCIETY AND HUMAN BEINGS CONFERENCE COMMITTEE MEMBERS Alvaro Taveira, University of Wisconsin Whitewater, USA Amanda Reggiori, Universita' Cattolica del Sacro Cuore, Italy Andrea Marcante, Università di Milano, Italy Andrea Perego, Universita` degli Studi dell'Insubria, Italy Andrew Dearden, Sheffield Hallam University, UK Andrew Sloane, University of Wolverhampton, UK Anita Mirijamdotter, Växjö University, Sweden Anna Spagnolli, University of Padova, Italy Annelie Ekelin, Blekinge Institute of Technology, Sweden Antti Pirhonen, University of Jyväskylä, Finland Aron Larsson, Mid Sweden University, Sweden Barbara Craig, Victoria University, New Zealand Barrett S. Caldwell, Purdue University, USA Bernhard Ertl, University der Bundeswehr, Germany Bettina Pedemonte, CNR/ITD, Italy Bill McIver, National Research Council, Canada Birgitta Bergvall-Kåreborn, Luleå University of Technology, Sweden Birgitta Kopp, Ludwig-Maximillian University Munich, Germany Caroline Park, Glasgow Caledonian University, UK Caroline Parker, Glasgow Caledonian University, UK Catherine McLoughlin, Australian Catholic University, Australia Charlotte Wiberg, Umeå University, Sweden Christian Licoppe, Telecom Paris Tech, France Christina Mörtberg, University of Oslo, Norway Claire Dormann, Carleton University, Canada Darek Haftor, Luleå University of Technology, Sweden David Hakken, Indiana University, USA xiii

Diane Whitehouse, The Castlegate Consultancy, UK Eduardo Villaneuva, Pontificia Universidad Católica del Perú, Perú Eila Järvenpää, Helsinki University of Technology, Finland Eleni Berki, University of Tampere, Finland Elisabetta Robotti, Istituto Tecnologie Didattiche, Italy Ellen Brox, Northern Research Institute Tromsø, Norway Eloy Irigoyen, E.T.S.I. UPV - EHU, Spain Elspeth McKay, Royal Melbourne Institute of Technology (RMIT), Australia Elvis Mazzoni, University of Bologna, Italy Enda F. Fallon, National University of Ireland Galway, Ireland Eran Fisher, University of Pennsylvania, USA Eric Maurincomme, Agfa HealthCare, Belgium Eva Jansson, Royal Institute of Technology, Sweden Fabio Dovigo, University of Bergamo, Italy Fernando Ferri, CNR - IRPPS, Italy Francesca Alby, Università di Roma, Italy Franci Pivec, Institute of Information Science, Slovenia Frank Teuteberg, University of Osnabrück, Germany Georg Hodosi, Stockholm University, Sweden Gianluca Miscione, International Institute for Geo-Information Science and Earth Observation, The Netherlands Graeme Johanson, Monash University, Australia Gudrun Wicander, Karlstad University Sweden Havard Bell, SINTEF, Norway Helena Karasti, Oulu University, Finland Homa Bahrami, Berkeley University, USA Ilaria Lombardi, Università degli Studi di Torino, Italy Ilkka Arminen, University of Tampere, Finland Jacob Palme, Stockholm University & KTH, Sweden Jacques Steyn, Monash University, South Africa Jan Meyer, Monash University, Australia Jane Lessiter, University of London, UK Jessnor Elmy Mat Jizat, RMIT University, Australia John Soren Pettersson, Karlstad University, Sweden John Waterworth, Umeå University, Sweden Jonathan Freeman, University of London, UK Jorge Marx Gómez, University of Oldenburg, Germany Juha Lindfors, University of Oulu, Finland Jutta Weber, Technical University Braunschweig, Germany Kai Kimppa, University of Turku, Finland Kalle Jegers, Umeå University, Sweden Karin Hedström, Örebro University, Sweden Karmele Lopez de Ipiña, Universidad Del Pais Vasco, Spain Katarina Lindblad-Gidlund, Midsweden University, Sweden Katy Buckner, Napier University, UK Kirsi Lainema, Turku School of Economics, Finland Larry Leifer, Stanford University, USA xiv

Larry Stillman, Monash University, Australia Lars Ilshammar, Labor movement Archives and Library, Sweden Laura Helle, University of Turku, Finland Lazar Rusu, Stockholm University, Sweden Lech Zacher, Kozminski Business School, Poland Lorenz Hilty, EMPA, Switzerland Love Ekenberg, Stockholm University & KTH, Sweden Lynne Baillie, Glasgow Caledonian University, UK Lynne Hall, University of Sunderland, UK Lynne Humphries, University of Sunderland, UK Magdalene Rosenmöller, IESE Business School, Spain Manuela Repetto, Istituto Tecnologie Didattiche, Italy Marco Lazzari, University of Bergamo, Italy Marco Padula, CNR-ITC, Italy Marcus Breen, Northeastern University, USA Maria Cristina Matteucci, University of Bologna, Italy Maritta Pirhonen, University of Jyväskylä, Finland Marja Kankaanranta, University of Jyväskylä, Finland Marja Vehviläinen, University of Tampere, Finland Marjo Mäenpää, University of Art and Design Helsinki, Finland Mathias Hatakka, Örebro University, Sweden Mats Danielson, Stockholm University, Sweden Max Liang, National Taipei University of Technology, Taiwan Michael Friedewald, Fraunhofer-Institut für System, Germany Michelle Robertson, Liberty Mutual Research, USA Mikael Collan, Åbo Akademi University, Finland Mikael Wiberg, Umeå University, Sweden Mikko Jäkälä, University of Jyväskylä, Finland Minna Silvennoinen, University of Jyväskylä, Finland Nagasaka Itaru, University of International and Information Studies Mizukino, Japan Nestor Garay, Euskal Herriko Unibertsitatea, Spain Nicola Doering, Ilmenau University of Technology, Germany Päivi Fadjukoff, University of Jyväskylä, Finland Paolo Cottone, University of Firenze, Italy Pascale Carayon, University of Wisconsin-Madison, USA Patrick LAW, The Hong Kong Polytechnic University, Hong Kong Patrick Rau, Tsinghua University, China Patrizia Grifoni, CNR - IRPPS, Italy Penny Duquenoy, Middlesex University, UK Peppi Taalas, University of Jyväskylä, Finland Per Flensburg, University College West, Sweden Pertti Järvinen, University of Tampere, Finland Peter Hoonakker, University of Wisconsin-Madison, USA Pirjo Elovaara, Blekinge Institute of Technology, Sweden Pirkko Nykänen, Tampere University, Finland Ralf Isenmann, Fraunhofer-Institut für System (ISI), Germany Raquel Flodström, Karlstad University, Sweden xv

Reetta Raitoharju, Turku School of Economics, Finland Reima Suomi, Turku University, Finland Robert Pinter, Budapest University of Technology & Economics, Hungary Roberta Bernardi, The University of Warwick, UK Roberto Polillo, University of Milano Bicocca, Italy Rudi Schmiede, Darmstadt University of Technology, Germany Sakari Taipale, University of Jyväskylä, Finland Sangeeta Sharma, University of Rajasthan, India Sanna Salanterä, University of Turku, Finland Sara Eriksén, Blekinge Institute of Technology, Sweden Sarai Lastra, Universidad del Turabo, Puerto Rico Sari Tuuva-Hongisto, University of Joensuu, Finland Sebastiano Bagnara, International Telematic Universit UniNettuno, Italy Seppo Väyrynen, University of Oulu, Finland Stefano Levialdi,Carlos III University, Spain Steve Walker, Open University, UK Stylianos Hatzipanagos, Kings College London University , UK Susan Dray, Dray and Associates, USA Susan Turner, Napier University, UK Sylvie Gangloff, FMSH, France Tero Vartiainen, Turku School of Economics, Finland Timo Lainema, Turku School of Economics, Finland Toomas Timpka, Linköping University, Sweden Uiara Montedo, University of São Paulo, Brazil Victor Kaptelinin, Umeå University, Sweden Ville Harkke, IAMSR/Åbo Akademi University, Finland Vincent Duffy, Purdue University, USA Virve Siirak, Tallinn University of Technology, Estonia Wendy Fisher, Open University, UK Wendy Olphert, University of Loughborough, UK Wolfgang Hofkirchner, University of Salzburg, Austria

xvi

KEYNOTE LECTURE ICT FOR A SUSTAINABLE SOCIETY? RETHINKING THE “GREEN IT” DEBATE IN THE CONTEXT OF SUSTAINABLE DEVELOPMENT Professor Prof. Dr. Lorenz M. Hilty Head of Laboratory, Technology and Society Empa Swiss Federal Laboratories for Materials Testing and Research Switzerland

ABSTRACT One central challenge of our time is to reconcile intra- with intergenerational justice in a world of finite natural resources. The vision of solving this dilemma has been known as “sustainable development” since the World Commission on Environment and Development (“Brundtland Commission”) have published their influential report in 1987. What is the role of ICT in the context of sustainable development? Will these technologies help to develop sustainable modes of production and consumption, or are they just adding to the burden we put on nature? The basic assumption of the talk is that ICT is both part of the problem and part of the solution. I is therefore essential to govern the development and application of ICT with regard to sustainability in order to maximize the positive and minimize the negative impacts. Some examples (within and beyond the current “Green IT” debate) will show that ICT does not automatically lead to a sustainable society, but has a huge potential to influence the metabolism of society. Conceptual and methodological frameworks that can be used for analysis and governance will be presented, such as the Linked Life Cycle approach.

xvii

PANEL APPROACHES TOWARDS ICTS AND SOCIETY – THEORIES AND METHODOLOGIES Wolfgang Hofkirchner University of Salzburg

ABSTRACT The panel attempts at working out categories along which a preliminary structuring of the field of ICTs and Society regarding its theoretical and methodological approaches can be carried out. Partners of a recently established open network meet to discuss this item. KEYWORDS ICTs and Society, Theories, Methodologies.

1. INTRODUCTION For little more than the last ten years there has been a remarkable shift in the fields of Science–Technology– Society, Communication Studies, Computer Science and adjacent fields towards more recognition of the role ICTs play for the advent of Information Society (or whatever term may be used to depict the society to come that is said to deeply depend on the usage of ICTs). A field of overlapping and converging concerns like Internet Research, New Media Studies, Social Informatics, and so on, has emerged and has been given “infrastructure” by the establishment of a variety of research centers and academic positions within given disciplines. A network of researchers engaged with ICTs and society was established only recently. It is clear that this field is in a rather premature state-of-the-art. Thus the body of theories is yet to be elaborated as is the toolbox of methodologies. For example, No. 4, Vol. 21, of The Information Society (2005) addressed the as-yet-to-be-sharpened profile of this new field. The panel aims at providing room to reflect theoretical and methodological assumptions characteristic of the research in ICTs and society as well as issues that might be crucial for the future profile of the field.

2. CONTRIBUTIONS • • • • •

Wolfgang Hofkirchner, ICT&S Center, University of Salzburg: A Taxonomy of Theories about ICTs and Society. Alice Robbin, Rob Kling Center, Indiana University Bloomington: A Preliminary Inquiry into the Methodologies Employed in Research on ICTs and Society Eduard Aibar, IN3, Universitat Oberta de Catalunya, Barcelona: New Theories and Old Mistakes: A Critical Analysis of Information Society Conceptualizations from an STS Point of View Lorenz Hilty, Technology and Society Lab, EMPA, St. Gallen: ICT and the Issue of a Third Industrial Revolution Gunilla Bradley, School of ICT, Royal Institute of Technology, Stockholm: A Convergence Theory on ICT, Society, and Human Beings

3. CONCLUSION This is but one attempt among others at working out some categories for the structuring of the field.

xviii

Full Papers

IADIS International Conference ICT, Society and Human Beings 2009

BUILDING TRUST ONLINE: DO PRIVACY POLICIES HELP OR HINDER? Margaret Tan Nanyang Technological University

ABSTRACT Online trust is arguably an important condition for successful e-commerce. To win the trust of online consumers, most ecommerce websites post online privacy policies stating their data handling practices. In this study, a linguistics-analytical framework is used to analyse online privacy policies of websites from Singapore and Australia. The findings indicate that most policies are vaguely worded and lack clarity in the coverage of their data handling practices. Particularly striking is the use of ambiguous language regarding the sharing of personal data with third parties. Although the objective of privacy policies is to build online trust, it appears that the way privacy policies are currently written, they may hinder the growth momentum of e-commerce. Beyond its value in enhancing the perceived credibility of websites, the accreditation by reputable privacy seal bodies may appear to have a positive impact in building online trust of consumers. KEYWORDS Online trust, privacy policies, data protection, e-commerce transactions, online seals

1. INTRODUCTION Over the years, various types of online frauds have been committed on the Internet especially the use of one’s personal data. Such incidents undermine online trust. It is therefore important to assure customers that their personal information will be protected when they make transactions online as they often need to divulge personally identifiable information (e.g., name, address and credit card number for payment, etc). As such information can be sensitive; they must be protected from intruders and hackers’ unauthorised use. Imagine the financial risks when one’s credit card or financial data is stolen; also imagine the liabilities if one is impersonated in an unauthorised manner. The consequences for not protecting online customers can be enormous. It has been estimated that if a company breaches confidentiality of consumers’ personal data, about 20 percent of its affected customers will not likely do business again with the company, about 40 percent will likely consider ending their relationship with the company, and about 5 percent will likely hire lawyers to sue the company (Friedenberg 2006). Indeed, companies may face lawsuits for breaching consumers’ privacy. Sears, for instance, is reportedly facing a class-action suit seeking US$5 million in damages after exposing consumers’ purchase history on its Managemyhome.com website (Hollahan 2008). ChoicePoint, a data-broker of personal information, was fined US$15 million for selling personal data to some seemingly ordinary organizations which turned out to be identity thieves (Federal Trade Commerce 2006). Identity theft has been a major concern in e-commerce because the victims not only suffer financial losses but may also face criminal charges for offences committed by the identity thieves. Besides personally identifiable information, websites can also collect non-identifiable or aggregate information, such as click-through behaviour of online visitors. It has become a practice for websites to collect aggregated information to enable them to profile the consumers and formulate marketing strategies. As marketers like to adopt a personalised approach, there are drawbacks if the consumers’ privacy is not protected. People will be less likely to participate in online transactions once they have a bad experience of their privacy being breached (Gartner 2005). Thus, if companies do not address online privacy, they can run into consumers’ resistance. So, they assert their promise to protect consumers’ data by posting online privacy policies that describe their data handling practices. However, privacy notices are only effective if consumers read, understand and use the information contained therein (Milne and Culnan 2004).

3

ISBN: 978-972-8924-82-9 © 2009 IADIS

2. LITERATURE REVIEW AND CONCEPTUAL FRAMEWORK In the extant literature, trust is defined as “the belief that a party’s word or promise is reliable and that a party will fulfil his/her obligations in an exchange of relationships” (Schurr and Ozanne 1985) and it exists when one party has confidence in the other party’s reliability and integrity (Morgan and Hunt 1994). In other words, trust pertains to the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor or control that other party (Mayer, Davis and Schoorman 1995). In today’s information society, consumers generally understand that it is not possible to eliminate all risks in e-commerce and that they also have the responsibility to exercise requisite caution and control over the use of their personal data when they make transactions online (Hsu 2006; Tavani 1999). Reading privacy notice is, however, only one element for consumers to understand and manage the risks of disclosing personal information (Milne and Culnan 2004). It is also important to explore how other complementary elements of the website can enhance credibility on protecting consumers’ privacy.

2.1 Information Adequacy This study adopts Pollach’s (2007) framework that focuses on how online privacy policies can establish online trust. According to the framework, shown in Figure 1, it is assumed that online trust can be established with consumers when websites provide comprehensive information of its data protection policy. Also, the websites must appear to have the credibility in terms of its compliance with information security standards and regulations. Pollach (ibid) argues that providing knowledge to consumers should be in the way of delivering a comprehensive scope of its data privacy that responds to five privacy concerns - data collection purpose, third party data collection, data storage, data sharing and marketing communication.

Figure 1. Framework on Online Privacy and Trust-building

Among the concerns, it appears that data sharing has been identified as the most prevalent among Internet users (Miyazaki and Fernandez 2000). The other significant concern relates to the extent to which consumers are able to control the use of their personal data. It is an important factor in building online trust because the process of data collection and its use by the company may involve other parties beyond the corporate boundaries. Control can be given in two ways: the choice to opt-in or opt-out. Opt-in strategy compels affirmative steps by consumers to allow collection or use of data, while opt-out requires affirmative steps to prevent collection or use of data (Bouckaert and Degryse 2006). Lai and Hui (2006) noted that the provision of opt-in and opt-out can be carried out through various configurations involving different wordings (positive versus negative statement) and checkbox defaults (pre-checked 7 as agreeing with the choice statement versus unchecked as disagreeing with the statement). Because on-the-spot decision makers tend to agree with the statements or defaults presented to them (Bellman, Johnson, and Lohse 2001), the wordings and checkbox defaults can significantly influence consumer choice with resulting implications for the longterm building of online trust.

4


2.2 Language Clarity As language serves to communicate about reality or expresses a speaker’s version of reality (Halliday 1970), it is critical to investigate how websites describe the corporate data handling practices. In this context, critical linguistics (Fowler et al 1979) that provide an analytical approach to inquire into relations of words and meanings deriving from string of words may be useful in the analysis of privacy policies. For instance, • Lexical choice: is the thoughtful selection of words in order to cover negative aspects and draw attention to positive features. • Nominalisation: is the substitution of active voice with nominal expression to attenuate negative sense in privacy invasion and de-emphasise who is responsible. • Passivisation: is the use of passive voice to conceal who is responsible for a particular action. • Negation: is the uplifting attention to companies’ refusal over unethical practices, such as selling consumer data, to insinuate credibility. • Modality: is the use of modal verbs and adverbs to downplay the unfavourable practices, such as data sharing or unsolicited emailing, and to neutralise negative feeling caused by such practices. Based on these parameters, Pollach (2007) examined the text and content of the data privacy policies by analysing what the policies do or do not mention regarding data handling practices and how these companies describe their practices. Four broad patterns of rhetoric features were found as follows: • Mitigation: downplays the frequency of a practice, eg. from time to time, sometime, occasionally. • Enhancement: is the emphasis on positive attributes, e.g, reputation or membership benefits. • Obfuscation: is the masking of particular practices or facts. • Omission of agents: is the elimination of active subject to hide who is responsible or involved. It is not surprising to note that many studies have shown that Internet users rarely use online privacy policies to alleviate privacy concerns (Hochhauser 2001; Schwartz 2001). This is because privacy policies appear to have been written in a mystifying manner, creating an impression that authors of the policies do not want Internet users to read them (Schwarz, ibid). Unfortunately, many policies do not address the sharing of personal data which most users are concerned about (Antón et al, 2004; Earp et al, 2005).

2.3 Perceived Credibility The framework also suggests that consumers are more likely to undertake repeat transactions when they perceive the websites to be credible in protecting their privacy (Pollach, 2007). Such credibility may be indicated by the presence of a trusted privacy seal on the websites. Privacy seal is an accreditation by a third party regarding the company’s data handling practices. The privacy seal bodies prescribe standards of data handling practices, and accreditation is contingent on company’s (continuing) compliance with the standards. Based on the framework, the objective of this study is to investigate the information adequacy and language clarity of online privacy policies with regards to the data handling practices of the websites in Singapore and Australia. The study also hopes to gain insights into the complementary role of opt-in/opt-out control mechanisms and privacy seal accreditations in helping to build online trust.

3. RESEARCH METHOD The data was collected from a sample of 100 privacy policy statements namely, 50 Singapore-based and 50 Australia-based websites. Singapore and Australia were used based on their international electronic readiness (e-readiness) rankings, and the wide usage of English language. E-readiness ranking of a country is a measure of its electronic business environment and is an indicator of how accommodating the country is to Internet-based opportunities. Both Singapore and Australia attained high rankings in electronic business readiness, reaching sixth and ninth globally in 2007 (Economist Intelligence Unit 2007). The websites were selected based on their commercial success using Alexa.com’s and Hitwise’s traffic rankings.

5

ISBN: 978-972-8924-82-9 © 2009 IADIS

3.1 Data Coding Each online privacy policy was coded using the coding schema shown in Table 1. The schema consists of 29 questions corresponding to the five key privacy concerns: data collection, third-party data collection, data storage, data sharing, and marketing communication. Table 1. The Coding Schema

Data collection

(1)Collection and storage of personally identifiable information (PII); (2)Collection of aggregate information; (3)Users’ ability to view and update data profiles; (4)Collection of users’ data via surveys; (5)Sweepstakes used to gather consumers’ data; (6)Obtaining users’ information from other sources; (7)Storage and usage of email address from inquiries; (8)Cookies; (9)Information on disablement of cookies; (10)Information on consequences of disabling cookies; (11)Web beacons.

Third-party data collection

(12)Types of data collected by third parties; (13)Third-party cookies or Web beacons; (14)Privacy agreement with third parties collecting data; (15)Opt-out of third party data collection.

Data storage

(16)Measures taken to ensure offline storage of data; (17)Measures taken to prevent unauthorized employee access; (18)Users’ ability to delete PII; (19)Records of PII kept after user deletes PII.

Data sharing

(20)Privacy agreements with business agents receiving PII; (21)Sharing of aggregate information with affiliates; (22)Sharing of PII with third parties other than business agents; (23)Sharing of PII with affiliates; (24)Sharing of PII with third parties other than business agents; (25)Selling of data; (26)Sharing of email addresses; (27)Sharing of data obtained in sweepstakes/surveys.

Marketing communication

(28)Unsolicited emails; (29) Unsolicited emails from third parties.

Source: Pollach (2007)

Each question in the coding schema was used to check against the content of the privacy policy examined. Questions were answered by condensing explicitly defined facts that could address each of the coding schema, rather than the general perception from the entire text. If a question could be answered by the schema, the coding item was considered “question answered”. When no information could be found in the privacy policy to answer the question, the coding would be written as a question mark (?), or called “question unanswered”. The number of questions answered was then calculated proportional to the total number of questions. Higher proportions of questions answered referred to higher levels of coverage of data privacy. The coding was done by two independent coders, and to ensure coding reliability, the policies were coded twice. The inter-coder reliability for Singapore was 97.3 percent with Cohen’s Kappa 0.95, while the rating for the Australia was 98.1 percent with Cohen’s Kappa 0.96. As the kappa coefficients were in the “very good” level (Landis and Koch 1977), the coding was acceptable for analysis.

3.2 Language Clarity Criteria Although online privacy policies should be written in an explicit manner with clear descriptions as ambiguity can cause confusion, yet often policy statements tend to use ambiguous terms or wordings that can present more than one possible meanings. In this context, the coding schema was designed to focus specifically on identifying the ambiguous terms and their frequencies. To understand the companies’ motives for presenting ambiguous description in the online privacy policies, the schema classified these terms according to their pattern of purpose. For instance, rather than tabulating the frequency of each ambiguous phrase found in all the policies, the average frequency of each pattern of purpose was calculated; which meant adding all ambiguous terms under the same pattern and then dividing the sum by 50, the number of online privacy policies examined per country. The calculation of frequency for each pattern is illustrated as follows: Average frequency of appearance for each pattern = (A1 + A2 + A3 + … + An) / 50 …where: A1, A2, A3, …, An = frequency of appearance for each ambiguous phrase, and n = number of ambiguous terms in the same pattern. As an example, suppose there were three ambiguous terms (the sharing of, is shared, you receive) under the pattern “omission of agents”, frequency for each ambiguous phrase was found as: the sharing of: 6 times; is shared: 2 times; you receive: 2 times. The average frequency for the pattern “omission of agents” would

6


be calculated as - Average frequency of appearance for “omission of agents” = (6 + 2 + 2) / 50 = 0.2. A high frequency score means that the pattern is often used in the online privacy policies.

4. DATA ANALYSIS 4.1 Information Adequacy Table 2 shows the breakdown of questions answered per category. The findings show that Singapore websites addressed 39.4 percent of the question while the Australian websites addressed 43.4 percent. The proportion of questions answered was low for all categories. The low proportion of questions answered pertaining to third-party data collection could probably be that not every website permitted a third party to collect data directly from its website, thus, third-party data collection was not mentioned in these policies. Among the remaining factors, data storage appeared to be the least described. This could mean that companies were not being explicit about their data storage practices, including safeguards for records, prevention against unauthorised access, and policy for consumer data removal from their records. In brief, the findings suggested that most privacy policies did not contain adequate information on their corporate data handling practices. On closer analysis, Australian privacy policies answered more questions than Singapore websites in all categories. This could be due to the regulation adopted in the respective countries. For instance, in Australia, privacy is protected by the National Privacy Act which means that all Australian websites must comply with the Act. On the other hand, Singapore has no centralised legislation on data privacy; although it is acknowledged that Singapore has separate legislation dealing with protection of personal data (IDA, 2000) and the Spam Control Act (IDA, 2007). Singapore adopts a co-regulation approach through a home-grown privacy seal, TrustSg. Membership in TrustSg is voluntary and to be certified with the TrustSg, websites must comply with its Core Principles that cover the areas of disclosure, privacy, fulfilment, and best business practices, as well as protection of minors and the elderly (IDA, 2003). Table 2. Findings on Information Adequacy Category

Number of questions

Proportion of questions addressed (%) Singapore

Australia

Data collection

11

35.0

41.6

Third-party data collection

4

38.0

41.0

Data storage

4

24.5

29.5

Data sharing

8

49.0

47.3

Marketing communication

2

53.0

71.0

All categories

29

39.4

43.4

While data sharing has been the top concern among Internet users because of the inherent information security vulnerability and which is the very essence of how one’s privacy can be breached, yet, Australia’s National Privacy Act and Singapore’s TrustSg Core Principles do not guarantee complete protection of this aspect of consumer privacy. For instance, TrustSg states that disclosure of consumer data beyond the purposes of collection is permitted when ‘the disclosure is clearly in the interest of the individual’ and ‘it is impracticable to obtain the consent of the individual’ (ibid, 2003). On the other hand, Australia’s National Privacy Act states that personal data shall not be shared unless ‘the individual concerned is reasonably likely to have been aware’ of the disclosure of data (Privacy Commissioner of Australia 1998). While many websites did not specify detailed information about their practices in sharing personal data with other companies (see Table 3), they would still share consumers’ personal and aggregate data with affiliates (Singapore, 72 percent; Australia, 76 perccent) and third parties (Singapore, 72 percent; Australia, 74 precent) by means of exchanging or selling data, for example, in lucky-draws, lottery tickets or sweepstakes. This finding should be of concern as companies’ affiliates or third parties may not have the same level of data protection as the website with which consumers made transactions.

7

ISBN: 978-972-8924-82-9 © 2009 IADIS

Table 3. Corporate Practices in Sharing Personal Data. Sharing with affiliates (%)

Answer Yes

Sharing with third parties (%)

Singapore

Australia

Singapore

Australia

38

28

14

14

No

6

4

26

6

If authorised

24

44

44

58

No answer

32

24

16

22

4.2 Opt-In/Opt-Out Mechanism In terms of the opt-in/opt-out mechanism, the results showed that most websites did not offer the necessary power to consumers to control the collection, use and sharing of their personal data. In fact, even when such controls were provided, opt-out was being used more frequently than opt-in (see Table 4). Opt-out has an advantage over opt-in as it would elicit more positive responses from people, including the unconcerned (Lai and Hui 2006). However, it was argued that if consumers were unlikely to read the privacy policy of websites they visit or make transactions with (Milne and Culnan 2004), then it followed that they would also be unlikely to assert the opt-out choice even if they did not want their information to be shared or did not wish to receive marketing communication. Unfortunately, it appeared that most websites tended to take advantage of such consumers’ non-actions by assuming that they were willing to let their information to be disclosed to other parties or to receive marketing communication. The problem is that websites can ‘manipulate’ consumers’ non-actions to be “yes” responses simply by implementing the opt-out on their websites. Table 4. Frequencies of ‘Opt-in’ and ‘Opt-out’ Singapore (%)

Provision

Australia (%)

Opt-out from third-party data collection

14

22

Opt-in for marketing communication

4

20

Opt-out from marketing communication

30

44

Opt-in for third party marketing communication

0

6

Opt-out from third party marketing communication

8

12

4.3 Language Clarity Table 5 shows the four patterns of ambiguous terms used in privacy policies: mitigation, enhancement, obfuscation, and omission. Compared to Australian privacy policies, Singapore websites showed more frequent uses of ambiguous terms. This pattern could probably be due to the regulation in Australia to have ‘clearly expressed policies’ as stipulated in Principle 5 of the Privacy Act (Privacy Commissioner of Australia 1998). In contrast, there is no such mandate for websites in Singapore. In fact, Singapore websites, seal or non-seal, have no obligation to keep ambiguity out of their privacy policies. The Core Principles of TrustSg do not expressly require ‘clear expression’ – it simply requires companies to ‘make readily available information about its policies and procedures for handling personal data.’ (IDA, 2003). Table 5. Patterns of Ambiguous Terms in Online Privacy Policies. Pattern of purpose Mitigation (to downplay frequency) Enhancement (to emphasise qualities)

8

Ambiguous terms in online privacy policies

Average frequency in each online privacy policy Singapore

Australia

occasionally, from time to time, sometimes, at times

0.74

0.60

Data sharing: trustworthy, reputable, carefully screened (third parties)

0.36

0.30

Spam or marketing communication: of interest to you, beneficial to customers, special offer

1.30

1.20


Obfuscation (to disguise particular facts) Omission of agents (to hide who is involved)

Hedging claims: at our discretion, on a limited basis, we reserve the right to, including but not limited to

9.62

8.52

Obscuring causality: not without your consent, not … unless you give us the.., not unless we obtain consent from

0.36

0.22

the sharing of, is shared, you receive

0.40

0.24

The analysis found that ambiguous terms appeared most frequently when companies describe practices of data sharing (Singapore, 48 percent; Australia, 50 percent), and marketing communication (Singapore, 33 percent; Australia, 30 percent) with data collection having lower frequencies. It is disheartening to note that companies appeared to present ambiguous terms to obscure privacy intrusion of data sharing, hence luring consumers into receiving unsolicited marketing communication. It appeared that most websites seemed to use obfuscation excessively to hedge claims; hence they are likely to mask or obscure unethical practices, such as the disclosure of personal data to third parties, or even to protect themselves from privacy litigations.

4.4 Perceived Credibility Only ten out of the fifty Singapore websites displayed the TrustSg seal. In contrast, no Australian website sampled showed any privacy seal. Interestingly, websites with TrustSg seal appeared to have better information adequacy than non-seal sites – the findings showed that for each category of questions, seal sites were found to be relatively more comprehensive, having about 84.5 percent while non-seal sites were having very low proportion about 27.8 percent of questions answered. It appeared that TrustSg accreditation may have successfully encouraged members to formulate online privacy policies to address key privacy concerns more comprehensively. Also, to note is that ambiguous terms appeared less frequently in seal websites than in non-seal sites. Hedging claims, under the pattern obfuscation, emerged as the most frequent with average number of times, 10.6 times in every Singapore non-seal site as opposed to 5.9 times in every seal site. Again, the TrustSg accreditation process appeared to have motivated its members to draft relatively less ambiguous description of data handling practices. In short, the accreditation of TrustSg seal appeared to have a positive impact on the quality of members’ online policies in terms of information adequacy and language clarity.

5. CONCLUSION Although the objective of privacy policies is to build online trust, it is argued that the current state of privacy policies may instead hinder the growth momentum of e-commerce transactions. This study found that most online privacy policies examined have much deficiency in providing sufficient information to protect data privacy. These deficiencies may impede the building of online trust. Indeed, most statements lack clarity in their descriptions particularly in the use of ambiguous language regarding the sharing of personal data with third parties. Language ambiguity appeared to be a key failing in many of the policies. When describing data sharing practices, the authors artfully select words to highlight positive features and downplay negative aspects. Nominalisation appeared to be used to remove the negative aspects of privacy-invasive practices. Passive voice was used when describing data sharing with third parties. As it is more difficult to process negative sentences than positive ones, negation, such as ‘we will not sell your data’ is often followed with rhetorical hedging such as ‘unless otherwise stated’ - which gave authority to websites. From an ethical perspective, companies should not use or share consumers’ data except only when they have given a confirmatory response to authorise them to do so. Although the results showed that opt-out choice was more frequently implemented than opt-in, more important, many companies artfully used the trick of lexical choice to describe consumers’ control options. Websites often made it vague regarding whether permission was the result of opting-in or not opting-out by simply stating, for instance, ‘Your data will not be shared with anyone without your consent’ or ‘We will not let any third party send you marketing messages unless if you permit us’. Such subtle lack of transparency may hinder the building of trust in ecommerce even though it appeared that the consumers have a self-directed control mechanism for privacy protection.

9

ISBN: 978-972-8924-82-9 © 2009 IADIS

In summary, this study adds to the general body of knowledge regarding the role of privacy policies in establishing online trust by extending the investigation to websites in Singapore and Australia. However, as this research investigated only a sample of only 100 online privacy policies, generalisability should be treated cautiously. Further, investigation of the scope of data privacy within the text of online privacy policies cannot be interpreted as the actual scope of data privacy being exercised by website operators. This is because many anecdotal cases regarding the breach of privacy showed that companies may not dutifully follow their own privacy policies practices. As such, there is a need for further research to shed light on the relationship between the promises contained in the online privacy policies vis-à-vis the actual conduct of corporate data handling practices. To conclude, the fact that while websites have a duty to protect consumer privacy, it is also incumbent on the consumer seeking privacy protection to read each website's privacy policy to understand its data handling practices and make an informed choice regarding the extent of information disclosure. Indeed, the role of consumer knowledge and control over ongoing participation in ecommerce activities is essential to the preservation of consumer privacy (Rifon, Lacrose and Choi 2005).

REFERENCES Antón, A.I., JB.Earp, Q. He, W. Stufflebeam, D. Bolchini, & C. Jensen. (2004) ‘The lack of clarity in financial privacy policies and the need for standardization’, IEEE Security and Privacy 2, 2, 36-45. Bellman, S., EJ. Johnson, & GL. Lohse. (2001) ‘To opt-in or opt-out? It depends on the question’, CACM, 44, 2, 25-7. Bouckaert, J. & HA. Degryse. (2006) ‘Opt in vs opt out: A free-entry analysis of privacy policy, CESifo Working Paper Series 1831 Earp, JB., AI. Antón, L.Aiman-Smith, & W.Stufflebeam. (2005) ‘Examining Internet privacy policies within the context of user privacy values,’ IEEE Transactions on Engineering Management, 52, 2, 227-37. Economist Intelligence Unit (2007) ‘The 2007 E-readiness Rankings: Raising the Bar’, Federal Trade Commission (2006) ‘ChoicePoint settles data security breach; To pay $10m in civil penalties, $5m for consumer redress’, http://www.ftc.gov/opa/2006/01/choicepoint.shtm, accessed Mar 2008. Friedenberg, M. (2006) ‘The coming pandemic: no, not bird flu. Identity theft’, CIO, 19, 15, 1. Fowler, R., B. Hodge, G. Kress, & T. Trew. (1979) Language and Control, London: Routledge. Gartner (2005) ‘Gartner survey shows frequent data security lapses & increased cyber attacks damage consumer trust in online commerce’, http://www.gartner.com/press_releases/asset_129754_11.html, accessed March 2008. Halliday, MAK. (1970) ‘Language structures & language function’ in J Lyons (ed.) New Horizons in Linguistics, London: Penguin, 140-65. Hochhauser, M. (2001) ‘Lost in the Fine Print: Readability of Financial Privacy Notices’ Hollahan, C. (2008) ‘Online Privacy’s Call to Arms’, BusinessWeek, 8 January,. Hsu, C J. (2006) ‘Privacy concerns, privacy practices and web site categories’, Online Information Review, 30, 5, 56986. IDA (Infocomm Development Authority, Singapore) (2007) ‘The Spam Control Act Comes into Effect’, 8 June. IDA (2003) ‘TrustSg: Merchants’ Guide to Managing E-commerce Risk and Fraud’. IDA (2000) ‘Factsheet on Dotcomming the Private Sector’, 1 August. Lai, Y.L. & KL. Hui. (2006) ‘Internet opt-in and opt-out: investigating the roles of frames, defaults and privacy concerns’, in Proceedings of the ACM SIGMIS CPR Conference on Computer Personnel Research, 253-63. Landis, J.R., & Koch, G.G. (1977) ‘The measurement of observer agreement for categorical data’, Biometrics 33, 15974. Mayer, R, J.Davis, & FD.Schoorman (1995) ‘An integrative model of org trust’, Academy of Mgt Review, 20, 709-34. Milne, G., & MJ Culnan. (2004) ‘Strategies for reducing online privacy risks: why consumers read (or don’t read) online privacy notices’, Journal of Interactive Marketing 18, 3, 15-29. Miyazaki, A.D. & A. Fernandez. (2000) ‘Internet privacy and security: an examination of online retailer disclosures’, Journal of Public Policy & Marketing, 19, 1, 54-61. Pollach, I. (2007) ‘What’s wrong with online privacy policies?’ Communications of the ACM, 50, 9, 103-9. Privacy Commissioner of Australia (1998) ‘Information Privacy Principles under the Privacy Act 1998’. Rifon, N.J., R. Lacrose & SM Choi. (2005) ‘Your privacy is sealed: effects of web privacy seals on trust and personal disclosures”, Journal of Consumer Affairs 3, 2, 339-62. Schwartz, J. (2001) ‘Privacy policy notices are called too common and too confusing,’ The New York Times, 7 May. Tavani, H.T. (1999) ‘Privacy online’, Computers and Society, 29, 4, 11-9.

10


CHALLENGES OF THE DEVELOPMENT OF A KNOWLEDGE ECONOMY AT GRASSROOTS LEVEL: CASE STUDY OF A CULTURAL CENTER IN SENEGAL Anne-Marie Touru Department of Information Technology, University of Turku 20014 Turun yliopisto, Finland

Tuija Tiihonen HIS-unit, University of Kuopio P.O.Box 1627, 70211 Kuopio, Finland

ABSTRACT New innovations have brought improvements to the developing countries. However, digital divide exists even today between countries and inside countries. Idea of this paper is two-sided. First the goal is to study from literature how Finland and the World Bank try to close the digital divide. Then the objectives and the actions of Finland and the World Bank are analyzed from the perspective of the grassroots level, which is based on our case study in Senegal. From these bases we try to find out, how well poor people and their needs are considered by Finnish foreign ministry and the World Bank. KEYWORDS Digital divide, Developing countries, Information systems, Knowledge economy, Senegal

1. INTRODUCTION Nowadays the ability to create and use knowledge is presently more important than ever for the competitiveness and welfare of countries, because, information and telecommunication technologies (later referred as ICT), the globalization process as well as life, material and energy sciences have developed dramatically over the past decade (Aubert & Reiffers, 2004, p. 9). However, this development have not reached everybody everywhere in the world and new problem, called digital divide, has risen. Digital divide means that some people have good possibilities and capabilities to use ICT where as other have not. This gap can exists between information-rich and information-poor nations as well as inside one country between “rich and poor persons, racial majority and minority groups, men and women, young and old, disabled and nondisabled persons, and so forth” (Hongladaram, 2006, p. 24). Presently many developed nations and international organization are interested to bind the digital divide. One of these countries is Finland, who has published its policy for promoting ICT in developing countries in year 2005. To achieve the objectives mentioned in this policy Finland has decided to cooperate with the World Bank. Thus, in this paper we will study, how Finland and the World Bank tries to support development of the knowledge economy in the developing countries. The main goal of this paper is to analyze from the grassroots’ perspective Finland’s and the World Bank’s principles to promote knowledge economy in developing countries. This perspective is based on empirical study made in Finnish-Senegalese cultural center, which is a development project supported by Finnish foreign ministry. The study was made by following qualitative research methods and analyzed by using context analysis (Tiihonen, Korpela, & Mursu, 2006). From these bases we try to answer to our research question: How well challenges of the ICT adaptation found in our case organization are considered by Finland and the World Bank? We think this question is important because users and workers of our case

11

ISBN: 978-972-8924-82-9 © 2009 IADIS

organization are very poor Senegalese people, thus, they can are representing the poorest people in the world. Finland and the World Bank claim that knowledge economy should touch even them.

2. THE DEVELOPMENT OF A KNOWLEDGE ECONOMY It is believed that ICT can reduce poverty and promote sustainable development by improving people’s access to government, education and health (Cecchini & Scott, 2003; Kenny, 2001; Kenny, 2000), thus, it has been seen as an opportunity for African countries to leapfrog in their development. Developments in the use of ICT and in the ICT infrastructure are often considered to be part of the knowledge economy (or sometimes referred as information or knowledge society) development in the society. Hence, many countries and international organization support the efforts of developing countries to become knowledge economies. In this chapter we will analyze and discuss on the policies that Finland and the World Bank (later WB) have for promoting knowledge economy in developing countries.

2.1 Preconditions of the Knowledge Economy in Developing Countries Finnish development cooperation is mainly based on the Millennium Goals of United Nations. This applies to promoting ICT in developing countries as well. Thus, only when ICT projects can support other development goals, like reduction of poverty or offering education for everybody, they are endorsed by Finnish government. Its aim of is to integrate ICT to different sectors, like education, environment, health and government, in developing countries. In addition Finland wants to develop knowledge economy in a sustainable way and follow this development in its partner countries. Finland collaborates with the WB in its efforts to promote knowledge economy in developing countries; hence, the policies of Finland and the WB are based on same principles. (Ulkoministeriö, 2005.) According to Aubert and Reiffers (2004) the WB believes that knowledge revolution is at the same time a challenge and a possibility for the developing nations. If developing nations can enter to knowledge economy they will be able to satisfy their socio-economical needs. Thus, the WB has helped the developing countries to make investments in information technology both in the public and the private sector; thanks to this many countries have experienced growth in the telecommunication sector, increased productivity of the private sector, and enhanced delivery of public services (Navas-Sabater, 2002, p. 4). However, there are still many reasons, why knowledge revolution may not happen in developing countries (Aubert et al., 2004). That is why the WB has designed a framework which can be used to analyze the ability of a country to become knowledge economy. In this framework variables to measure the performance of the knowledge economy of a certain country are divided to four pillars. These pillars with variables inside them are (Aubert et al., 2004, p. 18): • The economic and institutional pillar: 1: tariff and nontariff barriers; 2: freedom to use alternative currencies; 3: property rights; 4: freedom of capital and financial exchange; 5: regulation of FDI; 6: regulation; 7: black market control; 8: voice and accountability; 9: participation of women in labor force; 10: control of corruption. • The innovation pillar: 11: researchers per 10,000 inhabitants; 12: FDI per 100 inhabitants; 13: trade (exports + imports) per 100 inhabitants; 14: science and engineering students (percentage of total students); 15: credit to private sector (percentage of domestic credit); 16: domestic credit provided by banking sector (percentage of GDP); 17: stocks trade turnover ratio (%); 18: market capitalization of listed companies (percentage of GDP). • The education pillar: 19: primary pupil-teacher ratio; 20: public spending on education (percentage of GDP); 21: years of education; 22: adult literacy rate (percentage aged 15 and over); 23: secondary gross enrollment ratio; 24: tertiary gross enrollment ratio (male and female); 25: tertiary gross enrollment ratio (female). • The ICT infrastructure pillar: 26: telephones per 1,000 inhabitants; 27: computers per 1,000 inhabitants; 28: Internet hosts per 10,000 inhabitants; 29: radios per 1,000 inhabitants; 30: daily newspapers per 1,000 inhabitants; 31: loss in electricity distribution (percentage of output). In this framework, as well as in many other WB publications (Dahlman, 2005; Guislan, 2005; NavasSabater, 2002), the development of the knowledge economy is thought to be depended on the economical,

12


institutional, educational and technological development of a country. Promotion of capitalistic principle and free market economy is thought to be highly important for development of knowledge economy, where as cultural factors and inequality between different social groups have not received almost any attention (Aubert et al., 2004). Also inequality between men and women gets only minor consideration because girls’ participation to primary and secondary education is not studied. These same principles can be found from Finnish policy as well. It declares that development in knowledge, skills, political-governmental situation, economical conditions and technical infrastructure are needed before than developing countries can become knowledge economies (Ulkoministeriö, 2005).

2.2 Discussion about the Preconditions of the Knowledge Economy Already a lot have been done for bridging digital divide between countries and inside countries, thus, the success of these efforts can be studied. Some researchers agree with the WB that capitalistic, free market economy with well educated people are preconditions for knowledge economy, however, some researchers and practitioners are saying that cultural and social issues should also be considered. In accordance to WB’s principles Kenny (2001) as well as Cecchini and Scott (2003) writes that competition among private telecommunication companies decreases the costs of communication services, which will lead to higher rates of the Internet, phone and mobile phone penetration. In commercially less attempting environments, like rural areas, communication and information services can be provided by small enterprises, which do not need as big profits as large telecom operators (Cecchini et al., 2003). In Senegal one successful solution has been private telecenters, which provide public access to the Internet with affordable price for poor people (Kenny, 2001). However, some regulatory mechanisms may be needed to provide ICT infrastructure even in the most distinguish areas. “One approach is to invite private operators to bid for services in areas that are not commercially viable in return for a subsidy financed from a universal access fund. A concession contract is then awarded to the company requesting the smallest subsidy” (Cecchini et al., 2003, p. 80). Bollou’s (2006) studies on, how Sub-Saharan African countries have succeeded in investing to ICT, partially supports the WB’s viewpoint to knowledge economy development. He noticed that ICT infrastructure in Côte d’Ivoire developed well, because it invested on ICT infrastructure constantly and significantly, made profitable deals with private telecommunication companies, had good civil infrastructure, developed ICT education and had low prices of ICT services. In Côte d’Ivoire investments in ICT infrastructure had also contributed to GDP growth. Similar success has not happened in Senegal, probably, because Senegalese road conditions are bad and possibilities for ICT education are limited. Among others, cultural and social conditions also affect, who can use ICT and how it is used in developing countries. Many researchers have noticed that gender roles and inequality between genders hinder women’s possibilities to use ICT (Hafkin, 2002; Huyer & Sikoska, 2003; Kennedy, Wellman, & Klement, 2003). For example Cecchini and Scott (2003) noticed that illiteracy and low level of education are powerful obstacles to the use of ICT in India. They think that in order to reduce poverty ICT projects must be developed with local people and answer to their information needs. In addition intermediaries, who are attached to community at stake, should be involved.

3. METHODS Research and analyzing methods used in our study are presented in this chapter. Our case study is somewhat special in the field of information systems because it analyses information system (later referred as IS) of an organization, which has poor technological infrastructure and largely illiterate personnel. Thus, many traditional frameworks were found inappropriate for this research; however, Tiihonen’s (2006) framework was found different. With this framework it is possible to analyze the context of the IS instead of the IS itself. This is a good quality for a framework, which is used to study ISs in developing countries, because they are not always based on ICT.

13

ISBN: 978-972-8924-82-9 © 2009 IADIS

3.1 Research Method Our case organization is a Finnish-Senegalese cultural center, which is situated in Dakar, the capital of Senegal. Our research in this organization was conducted based on qualitative research methods defined by Strauss and Corbin (1998), hence, hypotheses were not defined before entering research field and all research findings rose from the research data. It was only in later when the suitable framework was chosen for this study. We used only qualitative research methods of which the participative observation (see Mikkelsen, 1995) was most important. The first author lived and worked in the cultural center for one month in spring 2006. She helped in daily work, participated to all official meetings and organized her own project. In addition five semi-structured interviews (see Hirsijärvi, 2000) were made to workers and users of the cultural center. Most of the interviews were personal, but one was a group interview (Table 1). Interviews were stored by writing them down during the interview. The language used in interviews was French and quotes of this paper are translated to English by the first author. Table 1. Information about the interviewees Interviewee Interviewee A Interviewee B Interviewee C Interviewee D Interviewee E Interviewee F Interviewee G Interviewee H

Sex Male Male Male Male Male Female Female Female

Status in the cultural Worker Leader Worker Worker Worker Worker Leader’s family member Leader of women’s department

Age 20 to 25 years 25 to 30 years 20 to 25 years 20 to 25 years 25 to 30 years 20 to 25 years 15 to 20 yars 25 to 30 years

Form of interview Group Personal Personal Group Group Group Personal Personal

We used qualitative methods also for analyzing research data. First research data was analyzed by open coding (see Strauss et al., 1998) and then it was gathered to under several themes (see Eskola & Suoranta, 1998). Lastly, data organized in themes was studied based on context analysis, which is presented in the next chapter.

3.2 Description of the Context Analysis To analyze our research data we used new framework called context analysis (Tiihonen et al., 2006). It is designed especially for studying information systems in the context of developing countries. The concept of context is huge and very difficult to define or explain, hence the framework of context analysis is aimed to be a tool to help understanding the context. After Heeks (2008) the successful IT projects are led by hybrids that span the technical and organizational. However, these hybrids are quite rare; thus, this framework offers context map, which is aimed to help anyone to expand their hybrid viewpoint. The map would help to separate and divide different items in the IS context, to classify and divide different components/items in the IS, for specializing what kind of expertise is needed in that point on the map. For instance, if the item/problem is on the socio-political category, it is not probable that it can be changed inside the organization, but inside the organization is possible to innovate suitable strategies to adjust the situation, and even try to benefit from it. The framework of context analysis is consisted of three different maps: the analysis levels of contexts, the scopes of contexts, and the categories within contexts. These maps can be used together or separately, they are not any methodology but rather tools to be used to outline the studied context. Levels of context are based on the 2X4 IS analysis level model is first presented by Korpela et al. (2001), which was modified by adding fifth, global level. Levels of this modified version are global, societal, organizational, group/activity and individual. The scopes of context are chosen based on anthropologist’s definition of three context levels: cultural, historical, and immediate context (Tapaninen, 2005). However, at the bottom of the three anthropological levels of contexts context analysis includes on more, which is nature. Thus, it considers all advantages and problems that natural environment can cause, such as humidity, coldness, heat, dust, or animals.

14


Several environmental categories of factors have an influence on the context of IS. In this framework they are delineated to five main categories: socio-political environment, infrastructure, organization culture, economy, and human resources. The selection of these categories is based on the findings in literature and the second author’s experiences in work and research in information systems. For our research purposes we analyzed the research data only in three categoriesall five categories. Social-political situation is analyzed in a social level, in this case Senegal, from the viewpoint of history, culture and present. Economy and infrastructure is studied in the organizational level from the present perspective. Here, the organization in question is the cultural center and social context around it is the suburb of Dakar surrounding the cultural center. The organizational culture and human resources are analyzed in organizational level and group level from cultural and present point of view. In the group level of analysis we use a division of two groups of people, who work in the cultural center: Senegalese women and Senegalese men. We also recognized other groups, like Finnish women, but they are not studied more precisely in this paper.

4. RESEARCH Our case organization is a cultural center in Dakar, the capital of Senegal. It is a cooperation project between Finland and one suburb of Dakar. Its goal is to improve the standards of living and the environment in this suburb. Currently, the center provides working facilities for artists and crafts men, afternoon clubs for children and education for women. It also hosts visitors from Finland. These visitors usually organize projects to which local people can participate. The idea of these projects can be e.g. teaching children, empowering women, or improving surrounding environment.

4.1 Socio-political Environment Senegal is the most Western country of Africa. Because of its strategically important location it has been affected by European countries for a long time. Until 1960 Senegal was a colonization of France and its culture is affected by French culture still today. The official language of Senegal is French, but Wolof is the most spoken. Senegalese people belong to several religious groups, but most of them are Muslims. Fortunately all religious groups get along well and there are only some minor problems between different ethnic groups. On the other hand poverty, underemployment, problems of health care and lack of education cause lots of problems in Senegal. Since independence the underemployment has risen especially among men. Women’s underemployment has not risen as much because they can do small business, like food selling in the market place. Thus, women’s role as family’s care taker has emphasized. (Rosenlew & Zilliacus, 2003.)

4.2 Economy The cultural center is not a wealthy organization but on the other hand its expenses are also small. Most of the monthly funding comes from Finnish partner organizations. In addition the center gets extra money by selling products, such as clothes and ornaments, made in there. The cultural center uses money mostly for the the rent, salaries and daily goods. Usually its employees work for free but some of them can get a small salary. Senegalese women get money when they sell clothes that they have made. According to the leader of women’s department money is divided like this: Small amount of money is given to women right away and rest is given to me. I save the money for one year and then I share it with the other women. Also sewing materials can be bought with saved money.

4.3 Infrastructure The cultural center is located in a poor suburb of Dakar and it is affected by the infrastructure of this suburb. Electricity is available, but power failures happen few times a month. Electricity is used only in houses and not in the public areas such as streets or meeting points. In houses electricity is used for lights and electrical

15

ISBN: 978-972-8924-82-9 © 2009 IADIS

devises like televisions and CD players but not for cooking or warming purposes. Refrigerators can be found only in food stores. Like other houses the cultural center has electricity and running water. Cultural center has one old computer with no Internet connection. Thus, the telecentres of the suburb are very important for the communication between cultural center and its Finnish partner organization. Mobile phones are frequent in the cultural center, although, local women do not have personal mobile phones; instead the oldest woman of their family usually has mobile phone, which other women of the family may use. In addition suburb has three telephone booths which are used to call abroud.

4.4 Organization Culture Organization culture of the cultural center and culture in its immediate environment is somewhat similar to West African culture, which is collective, feminine, and has high power distance and low tendency to avoid uncertainty (Hofstede, 1992). Cultural center also has some characteristics which are not normal to culture in its immediate environment; it seems that these habits are learned from Western organizations. Organization culture of the center is very collective. First of all, no one is usually alone there. Secondly, cultural center is in close contact with the immediate community around it. Almost everybody who works in the center lives next to it. Their female relatives often come to help in the center, usually for cooking or cleaning. Also other people, who act in the center, often live nearby. Thirdly, the workers of the cultural center do not only work there, but they also spend their evenings there. They often invite their friends there too to have fun. Fourthly, most of the actions take place in the suburb surrounding cultural center, for example workers of the cultural center take care of the trees which they have planted to this suburb. Leader has very important role in cultural center. It is commonly understood, that he is responsible for everything that happens in or by the cultural center. Also every activity that takes place in cultural center or is done by cultural center must have his permission. That’s why all formal information have to be told first to leader who then he tells it to others, and if this order isn’t respected nothing will happen. In addition leader is also the coordinator who divides all the tasks and gives feedback on them. In his own words he does “nothing and everything”. It seems that women and men aren’t equal in cultural center which is in line to culture of surrounding community. In cultural center women’s inferior role is visible in two ways. First, women do not have administrative tasks except in women’s department; however, women’s department is largely independent from other parts of cultural center. Secondly women do not participate to decision making process of the cultural center. For example, they did not participate to those two meetings which were held during the Touru’s research period in February, and usually women do not speak during official conversations even if they are present. This became clear to Touru when she did a group interview to one female and three male worker of the center. During this interview woman did not say anything except if she was asked something directly, and even then she answered very shortly and one of the men continued to answering for her. This interview session was totally different from another interview, which was made the leader of women’s department. During this a lot was spoken even if the leader of women’s department could not speak French fluently. This problem was solved by asking younger women to translate. Even if women aren’t as respected as men in cultural center the old Senegalese women make an exception. They are treated with high respect by Senegalese men and younger women. Differences and uncertainty belong to life in cultural center. People are used to work with people who belong to different religion, nationality or ethnic group, for example, cultural center works together with Christian school despite the fact that most of its workers are Muslims. Because of uncertainty workers of the cultural center are used to think that everything does not always go as planned. This appears in the way that people speak because they end every phrase about the future with a word “insha’Allah”, which means if God wills. Previous characteristics of the culture in the center are largely similar to culture of the community around it. However, the cultural center has borrowed some ways of action from Western cultures: For example workers of the cultural center organize meetings regularly and write notes during these meetings. The way that people behave in these meetings is different than their normal way to behave; during the meetings people do not talk as lowed and freely than usually. Instead leader is chairing the conversation. Usually only the most respected members of the organization speak and others only answer to questions directed to them. Still

16


meetings have something common with local culture; these meetings are held very late in the evening at the same time that workers of the cultural center would be together anyway. Based on organizational customs, attitudes toward future, and leader’s role in the organization it seems that cultural center has collective culture with high power distance and low tendency to avoid uncertainty. This is not a surprise because these dimensions are common to West African culture as well (Hofstede, 1992). On the other hand Hofstede’s (1992) claim that West African culture is more feminine than masculine does not seem to apply to cultural center because women’s and men’s roles and responsibilities are remarkably different.

4.5 Human Resources Because women and men are unequal in Senegal, in the cultural center they have different kinds of skills. In In women’s group illiteracy is common, and their French skills are poor. Furthermore computer literacy is poor, thus, it is hard to use computer or the Internet if one cannot read. However, they are able to learn using computer by memorizing different icons. Thus, with the help of literate person they could use computer and the Internet quite well. In any case during the research period of the first author Senegalese women did not use e-mails or text messages for communication. But, most of them had a possibility to make mobile phone calls. In men’s group French speaking skills were good, where as abilities to read and write varies a lot. Some of them read and write very well, but some of them read and write only poorly. Same thing applies to computer literacy: Some male workers of the center can use computers, the Internet, MS office programs and e-mail services but not all of them. Those who have good computer literacy can also read and write. Thus, they use e-mails and text messages, but, the most popular way to communicate in men’s group is to speak or call to others. Even if most of the Senegalese women and men did not have good education they respected knowledge. For example leader of the cultural center said that “the most important thing is… to know more about the world around us. It would benefit the inhabitants of the suburb”. Hence, the Internet was well known there, even if most of them had not used it. They knew that Internet is a good way to get different kind of information and communicate with people around the world. Workers of the center thought that their organization could become well known through the Internet as said by the leader of the cultural center: The most important thing is to make center and its organization well known. It could be done easily in Internet. Through Internet small community and its activities would became better known and easily accessible all over the world. The individuals’ skills affects their role in the organization. Men who have good literacy and good computer literacy had more power and responsibilities than those who cannot read, write or use computers. This is understandable because administrational tasks of the center demand reading, writing and computer skills. Thus, it can be that the difference between the skills of Senegalese women and men is one of the main reasons, why women do not have as much power as men in the center. If this is true then it is possible that in future women will have bigger role in the center. We believe to this because many young women or girls living in this suburb can speak good French and have good reading and writing skills. Some of them can also use computers and Internet. If one day they will work in the center, they have ability to participate to administration and decision making of the center.

4.6 Results Based on our research it seems that there are many problems to solve before everybody in our case organization can use ICT. ICT development of this organization is challenged by illiteracy, lack of language skills, poverty and poor power-distribution network. In addition inequality between members of the organization affects their possibilities to use ICT. In our case organization women do not use as much ICT as men because they have inferior position in the case organization and the surrounding community. Since women are less respected than men in the community, they typically cannot read and write as well as men. In addition they do not speak French as well as men. Thus, they have difficulties when using computers and the Internet. It is also possible that their use of ICT has not been considered important because of their inferior position in the organization.

17

ISBN: 978-972-8924-82-9 © 2009 IADIS

There are also differences between the men in their abilities and possibilities to use ICT. Those men who have larger organizational responsibility also have best reading and writing skills. Because the leader of the cultural center is responsible for all decision making and information sharing in the organization, his use of ICT is seen more important than the use of ICT by other members of the organizational. Hence, it seems probable that most respected members of the organization will be the first ones to use ICT. Extension of ICT use to all workers of the cultural center would be very slow if current situation continues. In our case community men’s unemployment is frequent. This means that families have to survive mostly on women’s income, which can support only basic needs. In addition men of this community are usually responsible for buying new equipments. Thus, when men are unemployed, the family does lack money and willingness to buy ICT. Hence, supporting men’s possibilities to work could improve family’s capability to acquire new technologies. Despite these problems we believe that in future ICT will be used more and by a larger group of people in our case organization and the community around it. This belief is based on a development which has happened in the community. In this suburb young people, boys and girls, are more educated and use more computers than people who work in the cultural center at the moment. It is possible that the new generation will have better opportunities to use ICT regardless of their gender or their social status. Thus, inequality between sexes and social statuses may decrease in the cultural center if these youngsters will one day work there.

5. DISCUSSION The WB recognizes that digital divide can exist between countries and inside countries (Navas-Sabater, 2002, p. 4). However, in the publications of the WB (Aubert et al., 2004; Navas-Sabater, 2002) digital divide inside one country often means digital divide between urban and rural areas or between rich and poor people. Our empirical study among poor urban people has shown that digital divide can also exist between neighbours, friends and family members. Thus, it is difficult to believe that only economical and infrastructural changes would remove the digital divide. In addition it would require cultural change. General illiteracy – and women’s illiteracy especially – in developing countries is studied in the framework of the WB (Aubert et al., 2004), however, the lack of language skills cannot be measured with it. Based on o it seems that language skills are an important requirement for computer and Internet use in Senegal because most computer programs and Internet sites are not available in their native languages. This is in accordance with previous findings about impacts of gender to use of ICT in Africa (Hafkin, 2002; Huyer et al., 2003). The biggest challenges in the approach of the WB to knowledge economy development appears to be that it underestimates the impact of culture on the development of a knowledge economy. Cultural aspects are difficult to find from WB’s publications about knowledge economy (Dahlman, 2005; Guislan, 2005; NavasSabater, 2002). Critique, which claims that the WB’s vision of development is economically biased, seems to be correct (Tammilehto, 2003). Similar criticism is partially accurate for Finnish policy as well. It is not interested in root causes digital divide neither it considers social and cultural preconditions for universal use of ICT. Instead it focuses on describing possible positive effects of implementing ICT to developing countries. Most important limitations of our study short research perioid, limited amount of interviewees and only on case organization. Thus, our results cannot be generalized to all Senegalese organizations. However, our research implies that culture can have an important role to play in knowledge revolution. Because of cultural reasons members of our case organization and case community had different kind of possibilities to use ICT. Thus, our study suggests that wide use of ICT would be predicted by cultural change in our case organization and case community. This statement sets a foundation for more ethical questions about digital divide and knowledge economy development. Is creation of knowledge economy an acceptable reason to change culture? Should Western people change culture in developing countries in order to reduce digital divide? Similar questions have already been studied in development studies (Escobar & Chase, 2002; Shiva, 1988; Tammilehto, 2003), but these question would be an important topic for further research in the field of information technology for development.

18


REFERENCES Aubert, J.-E. & Reiffers, J.-L. 2004. Knowledge Economy in the Middle East and North Africa: New Development Strategies: World Bank Publications. Bollou, F. 2006. ICT Infrastructure Expansion in Sub-Saharan Africa: An Analysis of Six West African Countries from 1995 to 2002. EJISDC, 26(5): 1-16. Cecchini, S. & Scott, C. 2003. Can information and communications technology applications contribute to poverty reduction? Lessons from rural India. Information Technology for Development, 10: 73–84. Dahlman, C. 2005. India and the Knowledge Economy: Leveraging Strengths and Opportunities. Herndon, VA, USA: The World Bank. Escobar, A. & Chase, J. 2002. The Spaces of Neoliberalism: Land, Place and Family in Latin America: Kumarian Press Incorporated. Eskola, J. & Suoranta, J. 1998. Johdatus laadulliseen tutkimukseen (2 ed.). Jyväskylä: Gummerrus Kirjapaino Oy. Guislan, P. 2005. Connecting Sub-Sahara Africa: A World Bank Group Strategy for Information and Communication Technology Sector Development. Herndon, VA, USA: The World Bank. Hafkin, N. 2002. Gender Issues in ICT Policy in Developing Countries: An Overview. Paper presented at the United Nations’ Expert Group Meeting on "Information and communication technologies and their impact on and use as an instrument for the advancement and empowerment of women", Seoul, Republic of Korea. Heeks, R. 2008. ICT4D 2.0: The Next Phase of Applying ICT for International Development. Computer, June 2008: 2633. Hirsijärvi, S. 2000. Tutkimushaastattelu: teemahaastattelun teoria ja käytäntö. Helsinki: Yliopistopaino. Hofstede, G. 1992. Kulttuurit ja organisaatiot. Juva: WSOY. Hongladaram, S. 2006. Information Technology Ethics : Cultural Perspectives: Idea Group Publishing. Huyer, S. & Sikoska, T. 2003. Overcoming the Gender Digital Divide: Understanding ICTs and their Potential for the Empowerment of Women. Instraw Research Paper Series no. 1, April. Kennedy, T., Wellman, B., & Klement, K. 2003. Gendering the Digital Divide. IT & Society, 1(5): 72-96. Kenny, C. 2001. Information and Communication Technologies and Poverty, TechKnowLogia, Vol. July/August. Kenny, C. J. 2000. Expanding Internet Access to the Rural Poor in Africa. Information Technology for Development, 9: 25-31. Korpela, M., Mursu, A., & Soriyan, H. A. 2001. Two Times Four Integrative Levels of Analysis: A Framework. Paper presented at the IFIP TC8/WG8.2 Working Conference, Boise, Idaho, USA. Mikkelsen, B. 1995. Methods for Development Work and Research – A Guide for Practitioners. New Delhi: Sage Publication. Navas-Sabater, J. 2002. Telecommunications and Information Services for the Poor: Toward a Strategy for Universal Access. Washington, DC, USA: World Bank Publications. Rosenlew, A. & Zilliacus, M. 2003. Diina Baax – Kyllä se siitä. Helsinki: Finlandssvenskt filmcentrum. Shiva, V. 1988. Staying Alive: Women, Ecology and Survival in India. New Delhi: Kali for Women. Strauss, A. & Corbin, J. 1998. Basics of Qualitative Research (2 ed.). Thousand Oaks, London, New Delhi: Sage Publications. Tammilehto, O. 2003. Globalisation and Dimensions of Poverty, Elements of Discussion: Ulkoministeriö. Tapaninen, A.-M.; Sosiaaliantropologian perusteet; http://www.avoin.halsinki.fi/Kurssit/sosAntr/materiaali; 4.5.2005. Tiihonen, T., Korpela, M., & Mursu, A. 2006. Creating a Framework to Recognize Context-Originated Factors in IS in Organizations. Paper presented at the IFIP-TC9 'Relationship between Computers and Society'. Ulkoministeriö. 2005. Tietoyhteiskuntaan sekä tieto- ja viestintäteknologiaan liittyvän kehityspolitiikan linjaus. Helsinki.

19

ISBN: 978-972-8924-82-9 © 2009 IADIS

EXTENDING A MEETING MODEL AND IDENTIFYING SOCIAL ACTIVITIES FOR APPLICATION SUPPORT Terence Blackburn, Kuan Tan, Paul Swatman University of South Australia 27-9 North Tce., Adelaide South Australia 5000

ABSTRACT One research area that has failed to produce any effective results is support for face to face meetings. In reality, there are almost no well known applications that support the inherently social nature of this type of group work. People in small teams do not always follow predefined processes, such as agendas, and they may ignore prescriptive direction. Given that people’s actions are unpredictable, it becomes very hard to provide technological support. Further, there is a lack of understanding of what teams really do during interaction and that meetings, in the holistic sense, are not being examined in enough detail. This paper has two main contributions. First, an extended model of meetings is presented that includes the cognitive system formed by team members as well as artefacts and resources. Next, we examine meeting activities at a fine grained level this includes some of the social aspects that accompany the more formal processes. We provide a mini case study to exemplify our model and provide evidence of social processes that need to be considered for support. KEYWORDS Meeting model, social processes, actions, activities, face to face, meetings.

1. INTRODUCTION We have seen little progress in recent decades for developing collaborative information technologies (CITs) that are not web based. Focal research domains include Group Support Systems (GSS), Computer Supported Cooperative Work (CSCW) and Groupware. One such area that has failed to produce any effective results is support for synchronous, colocated team work; in other words, face to face, meetings. In reality, there are almost no well known applications that support the inherently social nature of this type of group work. Researchers have been unable to leverage the computational benefits of software applications to support the ill structured nature of group interaction that emerges when teams engage in their work. Given that this area of research seems to have had almost no impact (Shumarova and Swatman, 2008), we claim that there is more that needs to be considered in a meeting than processes and agendas. Further, there is a lack of understanding of what teams really do during interaction and that meetings are not being examined at a suitable level of fidelity. Formal CITs that might be used to support teams in meetings impose a process (agendas may be useful as a guide), they can only operate within a defined structure and have a limited universe of discourse (Wegner and Godin, 1999). This does not match well with the way that meetings operate. In reality, people can ignore prescriptive direction, they have an unlimited universe of discourse and are not predisposed to adopt structure. Their actions are unpredictable, highly flexible, nuanced, and contextualised (Ackerman, 2000). The paths that people follow through meetings are emergent (Suchman, 1987) and contingent on factors such as other people, the task and the environment (Lave, 1988). How can we design support applications for team activities when even the team members are unable to predict what they might say or do during the task? When we turn to Requirements Engineering for clues about this problem (a common starting point in the design of an application), we discover that data capture methodologies are not always effective. Data capture techniques are often based on what users or researchers “think people do” as opposed to “what they really do.” Analyses of human activities are often informed by post hoc rationalisation and reconstruction of people’s actions as well as theories of cognition and Rational Choice (Scott, 2000). These approaches may not necessarily produce sound or accurate results (Ford et al., 1991).

20


Application support for face to face teams is one of the focal points for the CreWS project. We are using a real time observational approach through a multi modal infrastructure to record our lab meetings and we have noticed that there are many other actions and activities that occur in meetings that are potentially worth analysing and considering in application design We report two main contributions in this paper. We revisit the issue of what components need to be considered in a meeting support architecture and then we look in a finer grained manner at the human actions and activities that might be important within such a framework. We extend Powell’s (2004) Input-ProcessOutput (IPO) model to include: the distributed cognitive system formed by teams, as well as components such as resources and artefacts. With our meeting architecture established, we observe a group of four people during a day long meeting. We found evidence that supported our meeting model and, in addition to processes such as decision making and team building, we discovered a number of other activities. For example, all team members had undeclared agendas and would lie to keep them hidden. These agendas were subject to capricious change contingent on environmental factors, other group members and time pressures. This paper is structured as follows. In Section 2, we provide a brief overview of some of the research relating to meetings and then, in Section 3, we examine the sorts of components we think are important for the purposes of building support applications. In Section 4, we share our observations of a meeting that took place in a car during a tourist expedition. In Section 5, we argue for validating our meeting model and, finally, we conclude and discuss where we are going with this work

2. BACKGROUND Meetings have been studied in many domains and for many different reasons. The corpus is large and we can only identify some of the main threads in this paper. Often the motivation for meeting research has been to increase productivity, which has been measured across various dependant variables. For example, information quality, contribution equity and members’ satisfaction (Flanagin et al., 2004) was the focus of one research effort while meeting success was also measured over discussion quality, member dominance & status effects (Rains, 2005). The most common meeting dimensions are size, duration and location (Panko and Kinney, 1995) as well as time and location (Dean et al., 2000). Sub sets of these dimensions include partially distributed groups (Johnson et al., 1996) and virtual meetings (Geyer et al., 2001). Interactive and social aspects of meeting research includes: analysing formal meeting protocols (Zhang et al., 2003), using popular literature to illustrate meeting dynamics (McLeod, 1999) as well as a field that focuses on multi modal research (Nijholt et al., 2005). The theme with this work is to observe small groups through video cameras and microphones, while experimenting with different inferencing approaches. Examples include identifying group action sequences (Al-Hames et al., 2006) and abductive models of behaviour (Aiken and Paolillo, 2000). There are many other research initiatives with different themes but one salient point that emerges is that each project examines just one or two specific aspects of meetings. This is not a criticism; merely an observation. The problem is that we still have no clear “big picture” of what happens in meetings from the perspective of being able to clearly articulate everything that is happening at any arbitrary point in time. This is necessary if we are to build infrastructures that can dynamically observe and understand the actions and activities in a meeting in the same manner as humans yet without human involvement. In addition to brainstorming, decision making and the like, people are simultaneously maintaining communication channels, managing group interactions as well as dealing with peoples’ egos, unforeseen problems, changing circumstances and other issues. This goes beyond agendas and process execution, and a useful first step is to try and classify all of the relevant entities or components in a meeting.

3. MEETING COMPONENTS In this paper we define a set of classes to represent the meeting activities, entities and artefacts that might be important. Previous work in this area included Powell (2004), who modeled team work using the InputProcess-Output (IPO) paradigm (see Figure 1). The inputs in this model: technical, training (experience), culture and design, are brought to the team by its members. The processing can be characterised as task work

21

ISBN: 978-972-8924-82-9 © 2009 IADIS

and different types of human interaction and the outputs are also anthropocentric. Our model extends this framework and includes: the system that represents the team; artefacts as outputs (they link pre and post individual work to team work) and resources other than those that directly relate to people

Figure 1. Powell’s (2004) IPO model of teamwork

Our perspective also examines relationships between the components and the base classes in this model, as depicted in Figure 2, are: • a distributed cognitive system; • a group of people; • social factors; • processes and • resources and artefacts. The distributed cognitive system, comprising all transient and persistent representations, is an architecture that includes people, technologies and artefacts. It focuses on information flow, coordination and cognitive offloading. The properties in the system include: duration, (temporal properties of a meeting), synchronicity (events that occur over contiguous periods of time) and context. Context is important if meetings are interrupted or postponed and resumed at a future point in time. It contains state (up to the point of interruption) and records: the condition of artefacts, elapsed time, the number of people, their identities and other variables. An important aspect of meetings is the interaction and communication between people. Interaction protocols and turn taking for speech (Parker, 1988) must be negotiated between the individuals (Graham, 2003) and this can happen in an explicit manner or it can be done in a more sub conscious way. Social aspects of interaction, then, are important. Verbal communication must be delivered and received in a shared context for correct semantic interpretation. This context includes volume, pitch, annunciation and similar properties. Other important non verbal aspects of communication which exist in meetings include: gaze (Novick, 2005), body language, body position and emotional states (Liao et al., 2006). Terms such as processes, activities, and actions are widely used and can have different domain meanings. We consider processes to be high level repeatable courses of action and, accordingly, highly structured. Below them in a taxonomy are semi structured activities. These are still abstractions and include decision making, idea generation and goal formulation. At the lowest level are the actions that are combined to form activities. They include: speaking, pointing, writing notes, listening and others. These are ill structured and cannot be prescribed in terms of when and how they must be executed. They will be different for each activity. For example, the conversations required to execute any arbitrary brainstorming session will differ between sessions and cognitive input is required to achieve different outcomes. Tasks and sub tasks are concretisations of activities that produce outcomes and artefacts

22


Artefacts have been defined as “man-made objects”. We extend this to include documents, drawings, digital and conceptual objects, processes, plans and agendas; most of which originate in conversation before being transformed into their consequent state. We consider that almost everything produced by group members is an artefact and, once created, becomes a potential resource for the team

Figure 2. A model of the major classes and relationships in team interactions

The environment is the main sources of resources for a task. Resources can be: physical, such as desks, computers and white boards; conceptual, such as previous discussions, goal hierarchies (Wright et al., 2000) and memories (Antunes and Carriço, 2003); or digital, such as bandwidth and archived data. Equally as important as the classes in a small group meeting are the relationships between them as shown in Figure 2. The set of rules {R} that govern these relationships is described below • R1 Resources cannot be accessed directly by group members but can be available in a meeting workspace (eg information can be pushed onto a public screen by the system) through actions • R2 Processes (including activities and actions) can directly access resources. This means that a person has to be engaged in an action to use a resource • R3 Roles, which are a social construct, can manage processes and, indirectly, access resources • R4 Group members can assume any of a variety of roles • R5 Members can access processes directly without assuming an explicit or meaningful role • R6 Members of the group communicate with other members in either a synchronous manner or an asynchronous manner depending on their context • R7 Members of the group create artefacts but only some will be used as resources. For example when many versions of a document are created it is usually the final version that is important. People, then, assume roles throughout the meeting and these may be transient or permanent. Through these roles, people can undertake activities and access resources. During the process of trying to achieve their goals, the group’s activities produce artefacts, which in turn may become resources for the group. We present evidence for our model in our case study.

4. CASE STUDY This study details the activities of four tourists as they drove on a day trip from Bled across the local Alps, south to Trieste and back to Bled (see Figure 3). The data was historically reconstructed the next morning by way of a recorded interview with three of the travelers. They left Bled (point A in Figure 3) at 10.00 in a hire car for a scenic drive with no clear route in mind but a desire to see Trieste if time would allow. (Dinner was thought to be starting at 18:00.) They obtained a map from their hotel and asked the car hire people to suggest a picturesque route and to circle some of the interesting destinations. Carlo took the role of driver and the rest of the team discussed who would navigate. This wasn’t popular role but eventually Mary reluctantly agreed to it. The road from Bled led North and then

23

ISBN: 978-972-8924-82-9 © 2009 IADIS

West through mountains. Just out of Bled, they arrived at a turnoff to a scenic gorge; a popular tourist spot. Mary made a passing remark that it was apparently a beautiful place. She suggested a visit but, after some discussion, the idea was rejected. The travelers then continued heading towards the Alps.

Figure 3. The route followed by the travelers (adapted from http://www.ired.com/pix/map/slovenia.gif).

The journey through the pass in the Alps was very scenic and they stopped often to take photos. With the support from the others, Mary was now comfortable as the navigator although everyone helped at times. The roads were stressful as they were narrow and wound slowly up and down through the mountains. This caused the trip to take longer than planned and by 14:00 they were only in Kobarid; one third of the way to Trieste. At this point they discussed changing their plans and most suggestions included missing Trieste to arrive back in Bled on time. Hilary, a new member in the group, sided with Mary for returning early whereas Wolfgang remained quiet. Hoping to get to Trieste, Carlo proposed staying with the original route suggesting that dinner would not start before 19:00. The tension in the group was permanent now with each person having their own (partially declared) agenda. Eventually, a compromise was reached such that a decision on which route to take would be deferred until an important intersection at point C in Figure 3. Mary estimated the times between towns as they drove in order to keep people informed and aware of progress. As the travelers reached Nova Gorica, an argument occurred about where they were. Mary thought they had already arrived at Nova Gorica but Carlo disputed it. He was proven correct a few minutes later when they arrived at a large sign with the name of the town. Everyone was now hotly debating about whether to go to Trieste. Mary, motivated by guilt and wanting to be seen to be "doing the right thing," expressed subtly that she did not want to go in order for the team to get back to Bled in time. Hilary still supported this idea and Wolfgang was still sitting on the fence. Carlo argued that there was enough time left and that they should stick to the original plan. The group resolved, instead, that they would drive a little further than the intersection (at point C in Figure 3) and go to the top of the hill that overlooked Trieste. They sped along and reached the hilltop earlier than expected. After studying the map and considering their current speed, Mary suggested that they might make it to Trieste and back to Bled on time. This served to reduce the conflict that was very evident in the group. Due to the early arrival at the hilltop, Hilary suggested that they should actually go down to have a look around Trieste (at point D in Figure 7.3). Everyone agreed and they had fun taking pictures around the water’s edge before returning to Bled. Upon arriving at their destination at 18:45, the group found out that dinner was scheduled for 20:00.

24


5. VALIDATING THE MEETING MODEL During the interview next morning a richer picture emerged. To avoid conflict, all of the group members had tacitly exercised a communications protocol that included vaguely stating their personal goals. Carlo wanted to go to Trieste but didn’t always admit this. Mary definitely did not want to go but did not make this clear. Hilary wanted to do the right thing for the group and return to Bled but did not state this forcefully. Wolfgang was keen to go to Trieste but, not wanting to favour only one side, remained silent. Wolfgang and Hilary shaped their articulated position from their perceived views of the group, which did not reflect the individuals’ true views. They adopted their positions to avoid adding to the stress and, accordingly, they avoided a confrontation and eventually achieved an outcome that seemed to be satisfactory. We argue that the activities in the car are analogous to those in a face to face meeting; the type that might be held by software developers, emergency service planners or other professionals. By describing the driving activities in the context of our meeting framework, many similarities begin to appear Disregarding the location of the meeting, the emergent actions and activities were similar to those in a semi-formal, face-to-face meeting. There were agendas, brainstorming activities, idea evaluations, negotiations, goals and roles. Some roles were interchanged during the course of the trip, e.g. the map reader role rotated occasionally. Some emergent agenda items, such as the decision to visit a town were negotiated on the run. Most of the members had dissimilar views of the global agenda. A example involved deciding whether or not to visit Trieste. This was very important for one member, but two of the others were indifferent and could have easily driven past the city. Due to the existence of private, unexpressed agendas, visible behaviours and expressed opinions were sometimes inconsistent with each person’s goals. This adds a level of complexity when trying to observe and interpret what people are doing during interactions. Their itinerary (agenda) was unplanned except for visiting some specific towns on the suggested route and the goal was to return to Bled for dinner. Problems such as an unsuitable map, the unknown nature of the roads and the unknown time for dinner had a substantial influence on the actions, activities and interactions in the group. As the trip progressed, the people realised that, due to time constraints, they might not be able to accomplish all their goals (i.e. visiting all desired places). As a result, tension and stress became evident during the second half of the trip as conflicts between different agendas crystalised. When people are communicating in synchronous activities they assume roles. These may be transient, such as a pacifier or explicit and permanent, such as a driver. Some sort of role assumption is inevitable if people engage in interaction. As communications progress, social factors such as dominance and attitudes produce leaders, facilitators, coordinators or any of a range of other roles. In our case study, the driver and navigator shared the roles of leader and facilitator. (Did that mean they were also making a tacit grab for power? This was not revealed or sought for in the interview.) Hilary emerged as a pacifier and this was effective for managing the stress produced by the challenging roads and arguments. Others became cheerleaders, where they provided encouragement for Mary’s navigation work. The cheerleaders also became assistant navigators at times such as when relevant road signs were spotted. When the value of the pacifier role became obvious, the navigator joined in and provided humourous progress reports. The goal structures were also similar to those in some meetings. There were two initial goals and these swapped importance as the trip progressed. The dominant goal (for Carlo, at least) was to get to Trieste and the lesser goal was to return to Bled in time for dinner. This agenda was discussed at the start of the trip and revised during the day. Eventually, each member of the group developed their own version. These were only partially articulated and this caused some of the stress. Each member’s goals were based on their own values and self interests. The group had to reconsider their goals due to the time lost crossing the Alps and the time for dinner was unknown. Changing the agenda is also common during meetings. If a presenter or speaker goes beyond their allotted time to speak, agendas have to be revised. The concept of individual views existing along with a global view was important in this case. If the personal views had been successfully reconciled into a consistent global view, there would probably have been less stress during the day and the travelers would have had more pleasant outcomes. Towards the second half of the trip, one member suggested bypassing Trieste and going directly back to Bled. This view enjoyed some support in the group but the idea was never discussed with total honestly. All group members held back their real thoughts and only expressed views that were (partially) acceptable to the group. In effect, and for different reasons, they all told lies. This is not unusual. People tell lies for many reasons such as protecting others, avoiding confrontation, escaping censure or getting what they want (DePaulo et al., 2004).

25

ISBN: 978-972-8924-82-9 © 2009 IADIS

The driver at all times wanted to go to Trieste even though he appeared to compromise by agreeing to go as far as the city lookout. The navigator did not want to go to Trieste and was not happy about the compromise but didn’t say as much in deference to the driver. The other two members based their expressed views on the apparent positions held by other members. One member expressly wanted to go to Trieste and one did not. They changed their expressed views to conform to the apparent desires of the group and to maintain a balance. This behaviour emerged within the group due to time pressure and the need to juggle goals. There is enough evidence to allow us to compare the activities during the car trip with those in our model of face to face meetings. There were negotiations, decisions made, idea generating sessions and different examples of group development theories. As people, the travelers handled all of these activities. They did not need better decision or group support systems, although some negotiation support might have been useful, but they did not communicate honestly. This, we have claimed, is a social factor and we are unaware of any social support systems that are available to assist group members while they traverse the rocky grounds of social interaction while under pressure. We recognise that juxtaposing the activities in a car, as we have described them, with traditional face to face meetings has limitations. The seating arrangement, while useful in a car, is unnatural in the context of meetings. This constrained the two people in the front seat with regards to communication channels. The verbal channel was almost exclusively used during the negotiations and decision making processes. This precluded using any nonverbal channels, which also carry important information. For example, when Mary proposed early in the trip that the group should visit the gorge (just outside of Bled), Carlo and Hilary misinterpreted her desire as just a suggestion. The request was not reinforced by other channels such as the facial movements of smiling and widening eyes

6. CONCLUSIONS AND FUTURE WORK In order to design applications to support face to face meetings, we have, as a first step, presented a fine grained meeting model with a comprehensive set of components. We have claimed that social factors also need to be considered and we have provided evidence of actions and activities that are part of the social aspects of meetings. We observed these actions and activities during a one day road trip and these are analogous to those in some types of face to face meetings. While these meetings may not have the same set of social influences or diversity of personal goals, we have observed many of the component classes of our meeting model in the scenario: • a distributed cognitive system: four people communicating, interacting and mediating cognition • a group of people: the four travelers; • social factors: roles; power grabs and attempts to dominate; personal and global goals; lies; as well as compliance and conformity; • processes: idea generation, team building and managing emotions and • resources and artefacts: maps, signposts, clocks. The relationships between the classes are also evident and some examples follow. People access resources, sometimes while playing a role, when they are engaged in a process (R1, R2 and R3). For example, the navigator tries to direct the team by consulting a map and observing signposts. Any other person using the map and looking at signposts can also assume the role (R4) of navigator although their purpose may only be to update their personal information about the environment (R5). The navigator must communicate (R6) with the driver or to the whole group, and the planned route becomes one of the resultant artefacts (R7). This artefact may undergo changes and the old versions will become less relevant. These are the actions and activities that we seek to observe and support in a real time manner. The travelers spent a lot of time observing the other members in the party through their own multimodal (eyes and ears) sensor framework. They collected a record of all observable communication actions, modeled and abstracted the actions, linked them to their personal, experience based models of behaviour and interpreted the actions as behaviour within their own comprehensive frameworks. This is the same process we are following with our multimodal framework. We are currently collecting and storing data from observations of our own meetings as we build an infrastructure to support face to face activities without human intervention.

26


REFERENCES Shumarova, E. & Swatman, P. A., 2008, Adoption and Diffusion of Collaboration Information Technologies: Are We There Yet? InSyL Working Papers. eConference Bled, Slovennia. Wegner, P. & Godin, D., 1999, Interaction as a framework for modeling. Conceptual Modeling, LA, CA, pp 243-57 Ackerman, M., 2000 Intellectual challenge of CSCW: the gap between social requirements and technical feasibility. Human-Computer Interaction, Vol 15, No 2-3, pp 179-203. Suchman, L., 1987 Plans and Situated Actions, Cambridge University Press, Lave, J., 1988 Cognition in Practice, Cambridge University Press, Cambridge, Scott, J., 2000 Rational Choice Theory. In Browning, G., Halcli, A., et al. (Eds.) Understanding Contemporary Society: Theories of The Present. Sage Publications. Ford, K., et al., 1991 An Approach to Knowledge Acquisition Based on the Structure of Personal Construct Systems. IEEE Transactions on Knowledge and Data Engineering, Vol 3, No 1, pp 78-88. Powell, A., et al., 2004 Virtual teams: a review of current literature and directions for future research. SIGMIS Database, Vol 35, No 1, pp 6-36. Flanagin, A., et al., 2004 Group Performance and Collaborative Technology: A Longitudinal and Multilevel Analysis of Information Quality, Contribution Equity, and Members' Satisfaction in Computer-Mediated Groups. Communication Monographs, Vol 71, No 3, pp 352-372. Rains, S., 2005 Leveling the Organizational Playing Field--Virtually: A Meta-Analysis of Experimental Research Assessing the Impact of Group Support System Use on Member Influence Behaviors. Communication Research, Vol 32, No 2, pp 193-234. Panko, R. R. & Kinney, S. T., 1995, Meeting Profiles: Size, Duration and Location. Twenty-Eighth Hawaii International Conference on System Sciences., pp 1002-1011 vol.4 Dean, D., et al., 2000 Facilitation Methods for Collaborative Modeling Tools. Group Decision and Negotiation, Vol 9, No 2, pp 109-127. Johnson, J., et al., 1996, Supporting Partially Distributed Groups in Electronic Meetings. Twenty-Ninth Hawaii International Conference on System Sciences, pp 112-117 vol.3 Geyer, W., et al., 2001, A team collaboration space supporting capture and access of virtual meetings. International ACM SIGGROUP Conference on Supporting Group Work, Boulder, CO. pp 188-196 Zhang, J., et al., 2003, Mediating Electronic Meetings. 27th Annual International Computer Software and Applications Conference, pp 216-221 Mcleod, P., 1999 A Literary Examination of Electronic Meeting System use in Everyday Organizational Life. Applied Behavioral Science, Vol 35, No 2, pp 188-206. Nijholt, A., et al., 2005 Meetings and Meeting Modeling in Smart Environments. AI & Society, Vol No, pp 1-19. Al-Hames, M., et al., 2006 Multimodal Integration for Meeting Group Action Segmentation and Recognition, Aiken, M. & Paolillo, J., 2000 An abductive model of group support systems. Information & Management, Vol 37, No 2, pp 87-94. Parker, K. C. H., 1988 Speaking Turns in Small Group Interaction: A Context-Sensitive Event Sequence Model. Journal of Personality and Social Psychology, Vol 54, No 6, pp 965-971. Graham, C., 2003 A Model of Norm Development for Computer-Mediated Teamwork. Small Group Research, Vol 34, No 3, pp 322-352. Novick, D., 2005, Models of Gaze in Multi-party Discourse. CHI Workshop on The Virtuality Continuum Revisited, Portland, OR. Liao, W., et al., 2006 Toward a Decision-Theoretic Framework for Affect Recognition and User Assistance. International Journal of Human-Computer Studies, Vol 64, No 9, pp 847-873. Wright, P., et al., 2000 Analyzing Human-Computer Interaction as Distributed Cognition: The Resources Model. Human Computer Interaction, Vol 15, No 1, pp 1-41. Antunes, P. & Carriço, L., 2003, Modeling the Information Structures of Meetingware. Workshop de Sistemas de Informação Multimédia e Cooperativos COOP-MEDIA, Porto, Portugal. Depaulo, B., et al., 2004 Serious Lies. Basic & Applied Social Psychology, Vol 26, No 2/3, pp 147-167.

27

ISBN: 978-972-8924-82-9 © 2009 IADIS

CONTEXTS FOR COUNSELLING - OPPORTUNITIES BY COUNSELLING VIA TEACHERS’ WEB FORUM Jarle Sjøvoll Bodø University College

ABSTRACT This article is based on a research project whose aim is to implement blended learning in a teacher’s web forum as a tool for solving practical problems in elementary and secondary education. The project concerns professional counselling as a resource for practical problem solving. The web forum is built on a simple principle: asking one question - getting different answers. Teacher’s questions are based on their own definitions and explanations of the practical problems they experience. Leuphana Universitet (DE) and Bodoe University College (NO) are partners in this research. Counselling via the teacher’s web forum offers many advantages in comparison with traditional, face-to-face counselling. First of all I would like to focus on choice and flexibility. Working online provides several advantages for counselling in terms of the available choice of counsellors and “meeting” arrangements, independently of time and space. Users can choose an online counsellor for reasons independent of geographical proximity, such as theoretical or religious basis or specialisation. Clients can start by making up their minds on the question of what kind of service they would like. From the counsellor's point of view, all clients will be met in the same way, independently of where they live. Reflection in and on practice is carried out by teachers who reflect on their performance together with others. In these learning communities, teachers can experience not only individual self-efficacy and its professional limitations but also social and cooperative self-efficacy. The online service will thus also focus on the needs of the teachers for personal and social support. Counselling by means of this web forum gives teachers the ability to solve their own practical working problems in a better way by using the synergy that can be gained by coaching in the teachers’ forum. Some people find it easier to discuss personally-significant topics without another person physically present, and some clients feel more comfortable in the relationship with their web-counsellor, but an open web forum like Teachers’ Forum is not so suitable for discussing specific personal problems. Email counselling may be an alternative in such cases. Teachers’ Forum seems to offer possibilities in terms of helping teachers and schools to build knowledge; it is also possible to offer social support and guidance in matters of substance. Visitors can also access information regarding ethics and laws. The users themselves explain their problems and expect to get an answer to the questions they have put into the forum. Here we can find questions relating to a need for information, advice on methods, and help dealing with situations and relationships with pupils, parents and teachers. The web forum seems to have a potential for guidance concerning practical problems. KEYWORDS Blended counselling, teacher’s forum, web forum, problem solving, knowledge building, social support

1. INTRODUCTION It is a commonplace that different teachers at different schools confront similar difficulties, such as aggressive students and demanding parents, as well as conflicts with colleagues or supervisors during their every-day work. These situations can be very draining, if not health-threatening. Instead of turning to others for help, teachers rather tend to deal with their problems in private, creating the same, sometimes ineffective or insufficient solutions over and over again. This is where the concepts of a “virtual platform” and “blended counselling” become useful. (Krøger, H. & Reisky, A. (2004) Valiathan, P. (2002) Experiences in relation to counselling via the teachers web forum seem to be supported by experiences of counselling via email. The “Teachers’ Forum” offers teachers the opportunity to work in professional and cooperative learning communities without having to reveal their identity. They now have the chance to admit to difficulties, to ask questions anonymously and to experience the strength and advantages of a professional learning community. They will realise that others have to face the same or similar problems and have found a way to solve them. This may help ease the pressure of the involved teacher, thereby enabling her to deal with her problem in a

28


more professional manner. By helping each other via the virtual platform, teachers can also improve their problem-solving abilities, as well as their abilities to offer encouragement and professional advice. Teachers’ Forum is also an instrument to increase, bundle and preserve expertise and knowledge and to make it accessible and useful for others many times over. Professional teachers, and retired ones as well, can offer advice and their immense knowledge and expertise will not be lost for younger colleagues. By eliminating intermediaries and allowing counselling work to take place without leaving the workplace or home, online counselling offers an unusual level of privacy. For many teachers, online counselling can take place from the client's own workplace or home. It also means the absence of intermediaries. It is not necessary to speak with a receptionist to arrange appointments. Finally, in terms of the actual process of problem solving which counselling involves, working online provides special advantages especially when dealing with personal problems like stress, anxiety, fear and other health issues.

2. BACKGROUND OF THE TEACHERS’ FORUM Teachers are society’s most important early warning and control system. Even though teachers are not trained to solve most social problems, they should be able to detect them, reflect on their critical importance, and solve them either by themselves, by finding experts, or by taking part in the solutions as co-therapists. Many teachers are overly burdened by these problems and react with critical stress or burnout syndromes. We assume a dysfunctional relationship between the health of teachers and the psychological health of children and their chance of a good education. Teachers are at the same time actors and instruments of their interventions. In contrast to handicapped people who can drive a car with special features to compensate for their difficulties, the stresses and moods of teachers will directly impact on pupils and their learning. Nevertheless, teacher health has up to now not been a subject in teacher training and has not been part of the focus of efforts for improving the educational system! With respect to health and professional quality, teachers need easily-contactable, multi-professional systems of assistance and learning. Severely stressed teachers and other professionals in leading positions who feel helpless to solve their problems should be given information about stress-reducing strategies and about how to solve educational problems both by specialists of different professions and by colleagues. Teachers should be able to express their feelings and discuss their educational practice with colleagues without having to reveal their identity within a learning community. They need encouragement to fight for their health and for the quality of their professional practice. They should have the chance to admit to difficulties, to ask questions anonymously and to experience the strength and advantages of a professional learning community (“One for all and all for one.”). Through the forum, teachers will receive acknowledgement of and understanding for their difficulties. They will realise that others have to face the same or similar problems and that they have found a way to solve them. This may help to ease the pressure of the involved teacher, thereby enabling him to deal with his problem in a more professional manner. This is made possible by the use of the internet. By helping each other by means of the virtual platform, teachers can also improve their problem-solving abilities as well as their abilities to offer encouragement and professional advice.

2.1 User Demand If necessary a teacher can address counsellors by email. There is no need to timetable meetings with the counsellors or for hours to go by in descriptions of sorrows during a long counselling process. The forum offers several more advantages not only to its users, but also from a public point of view: - Opportunity to receive help just when it is needed - meaning whenever problems are pressing - Opportunity to get a variety of different solutions from colleagues and professionals within a short period of time without any costs or member fees - Opportunity to choose from amongst several solutions the one that best fits the person asking for advice - The user who is seeking advice needs to describe his problem in an understandable manner, thereby gaining a less emotional perspective onto his problem

29

ISBN: 978-972-8924-82-9 © 2009 IADIS

- Opportunity to help others by reporting about solutions that proved successful. Expensive experts can thus be used more effectively and an insight can be gained into the professional work of a teacher. From a long-term perspective it becomes obvious that a virtual platform for teachers can be seen as an instrument that will lead to an increase in teaching standards and to better practice, as well as to an improvement in teachers’ health and efficiency.

3. THEORETICAL REFERENCES Any pedagogical design, including designs of online CKB, (Collaborative Knowledge Building) implies a latent – and sometimes unconscious – theoretical perspective on what learning is and what it ought to be. Such perspectives always form the context for the choice of pedagogical elements characterising the learning process. Therefore, it seems essential not only to become conscious of these underlying perspectives, but also to base designs on a clear and conscious theoretical understanding, in order to be able to choose corresponding pedagogical-didactic features that are likely to promote the learning perspective and goals in question (Sorensen 2003 p. 25). The building up of knowledge occurs both as an individual and as a social process, in that the personal acquisition of knowledge happens whenever an individual interprets another’s utterance. An assertion will, according to constructivist theory, be interpreted in accordance with the background of an individual’s understanding of the topic, and will be taken in to the individual recipient’s system of concepts. Knowledge, therefore, is not “transferred” unaltered from one individual to another, but is mediated, and “construed” anew by the recipient. Mason claims that successful online-based learning must build upon a constructivist theory of knowledge. (Mason 2003, Imsen 1998, Lund 2005, Vygotskij et.al 2001) A social constructivist approach to learning understands it as a process between people. Vygotskij claims that “learning happens when the culture is incorporated with the person” (Bråthen 1996). Language is the central means of ensuring a shared cultural heritage, and understanding is handed on at individual level. In a constructivist perspective, knowledge is not a fixed commodity, but something that is construed in a meeting between people, with language as the medium. The social-constructivist approach emphasises that the construction of knowledge is not an individual but a collective process, with language as the bearer of the collective understanding. (Imsen 1998, Lund 2005) The socio-cultural approach emphasises that learning requires knowledge to be presented in an authentic context, i.e., settings and applications that would normally involve that knowledge. Learning requires social interaction and collaboration (McLelland 1995, Lave & Wenger 1990). On-line guidance as a phenomenon contains so many forms of collaboration that these cannot be located directly within traditional categories of counselling and guidance. However, it is interesting to look at online guidance in relation to different aspects to which traditional guidance-theory relates: instruction, therapy, consultation, and work-based, practice-based, and personal guidance, for example. (Caplan1970, Lauvås and Handal 1990, Schön 1983, Argyris 1992, Lund 2005). “Traditional” guidance theory, as exemplified by Lauvås and Handal, is also useful for understanding different forms of on-line guidance and advice. But this theory must be supplemented with knowledge of what characterises the on-line interaction, its possibilities and its limitations. Here we have new qualities which typify the text-based, a-synchronic communication. As an example of this, the text stands alone as intermediary, without the possibility of being corrected or nuanced by other forms of communication. It is also typical of this a-synchronic communication that a message cannot be adjusted, nuanced or amplified from the background of the recipient’s direct reaction, as can happen with synchronised forms of communication.

4. RESEARCH AND PROCEDURES The Teachers’ Forum has become an integrated part of a dual counselling, assistance and learning system developed by both research partners. We are now interested in the evaluation of user experiences by use of the ne-forum. The main research question concerns the usefulness of the activities at the Teachers’ Forum. The survey is designed to assess the quality of user activity at the Teachers’ Forum. It is also important to

30


know what visitors are doing at the Teachers’ Forum. This may be analysed by the use of text-analytical methods to examine the Teachers’ Forum. This study is restricted to being an evaluation of the experiences of users who are trying out this new platform for online teacher guidance. Problem statement: What are the attitudes and experiences of users when they seek advice online due to encountering problems at school? Research questions based on your experience as a user of the Teachers’ Forum: 1) Do you think you have acquired new knowledge from the forum? 2) In what ways do you think Teachers’ Forum has a role as social support? 3) How do you evaluate the professional guidance received through participation in this online forum? The sample of the study consisted of teachers who are listed as users of Teachers’ Forum. The list does not include visitors who just use TF for reading. More than 356 teachers have answered parts of the online survey, but only 99 teachers have done all parts of this survey. The results are based on responses until December 2008. A five-point Likert scale was used to get a picture of the attitudes. In addition to the three subjects we can give a presentation of some demographical data collected. The survey was developed partly by analysis of questions and answers put into the Teachers’ Forum, partly based on theory of counselling. The subjects are: - Knowledge building (with 11 indicators of attitudes) - Social support (with 11 indicators of attitudes) - Professional guidance (with 11 indicators of attitudes) Respondents had an opportunity to add indicators that they could not find in the list, but which they considered to be important. All items were answered on the traditional five- point Likert scale, with 1 indicating ”strongly agree” and 5 indicating ”strongly disagree”. The research material consists in addition of texts that can be found in the Teachers Forum (TF). These texts have been used in the process of developing indicators and validating the questions, in discussion with the professional counsellors. Here follows general information and user evaluation on Teachers’ Forum.

5. RESULTS 5.1 Demographic Information The evaluation group consists of 32 % men and 68 % women. User ages are: Younger than 20 years 1 %, 20 - 29 years 10 %, 30 - 39 years 26, 5 %, 40 - 49 years 23, 5 %, 50 - 59 years 31 %, 60 years and older 8 %. Teachers are the biggest user group (79%), but teacher students, career counsellors, administrators, consultants and social workers are also using TF. The teachers work at the following levels: Primary and secondary school (35, 5 %) and higher level (16-20 years) - 25 %. The users can be classified in three categories, based on answers from 222 informants: 1. Intensive users: Some of the users are visiting TF many times a week (8, 5%). They may be termed intensive users (19 out of 222 respondents). 2. Regular users: The 60% who are regular users can be divided into 18% 840 users) who visits TF many times a month, and 42% (93 users) who visits TF several times a year. 3. One-time users: Nearly 1/3 (31, 5%) have visited TF only once. More than 64% of the respondents have known Teachers Forum more than one year. Most of the users are reading the contributions and counsellors’ comments (176 of 296 respondents – 59%). But 28% have posted a first contribution and 27% have made written comments.

31

ISBN: 978-972-8924-82-9 © 2009 IADIS

5.1.1 User Evaluation User evaluation is based on responses from 131 persons and from a ranking of the most important functions of Teachers Forum. Ranking progression is based on answers in group 1 (agree very much) and group 2 (agree to some degree) at the Likert scale. Only answers with more than 50% responses in group 1+ 2 are registered as the most important functions in relation to the three main concepts shown below. Each concept was evaluated by use of eleven indicators. Knowledge building Knowledge generation Knowledge application Raising the quality of professional Reducing professional isolation Health promotion Practical reflection More co-operation

% 75 69 60

Social support Universal issues Identification Problem solving skills

% 92 82 79

Professional guidance Empathy Trust Self-value

% 80 64,5 64

60

Interpersonal learning

74

Inclusion

63

52,5 52 51

Discussion about solution Altruism Change Accepting

72 71 70 70

62,5 59 56,5 53,5

Speaking out about issues

54

Problem detection Self reflection Coping strategies Emergency assistance in problem solving Resource activation Relations

53,5 52,5

Registered users can write questions and answers or personal mails to the specialists and they can get a licence to become an official advisor. They receive help at the time it is needed (whenever problems are pressing) and they can get a variety of different solutions from colleagues and professionals within a short period of time without any costs or member fees. From amongst several solutions on offer they can choose the one that best fits their individual needs. Users seeking advice are trained in asking complete questions. They need to describe their problem in an understandable manner, thereby gaining a new and perhaps less emotional perspective over their problem; they can also help others by reporting about solutions that have proved successful.

6. DISCUSSION Some features of online-based counselling and collaboration are significantly typical. Participants bring forward questions formulated on the basis of their own working situation. A large group then enters into a “discussion” and it is seldom that one person by virtue of his position or competence is expected to give a certified “right” answer. Dialogue is typified more by questioning reflection over compound problems than by simple questions and unambiguous answers. There is more emphasis on the shared learning process than on the individual progress / learning acquisition of any single person. A social-constructivist perspective on learning seems to be a fruitful and relevant angle of approach to take up the main sides of the phenomenon of online-based learning. Schön (1987) pointed out that the working out of a task often involves ways of posing problems which are ambiguous, containing uncertainties, unknown elements and possible value conflicts. The situation cannot be simply defined from training and earlier experience. In such situations, knowledge gained from previous work-experience is inadequate. Even for unregistered guests, Teachers’ Forum offers special benefits. Visitors can read case studies and proposals to help obtain ideas on how to deal with similar problems. By seeing that the problems discussed in the forum are more severe than their own problems they can gain increased confidence in their own capability. They can send case studies to colleagues to help them in their troubles and they can experience a learning community with the attitude “one for all, all for one!” Our evaluation has shown that the users of TF appreciate the web forum most of all because of the universal issues that are taken up and because they can identify themselves with the person explaining the problem. The web forum is generally a good tool for social support and personal guidance. Teachers’ Forum

32


also accumulates knowledge and disseminates such knowledge, especially for visitors. Users also take part in building knowledge. Writing about the problem is a part of the process of problem solving. For most people, typing is harder and more time-consuming than speaking. Because it “costs” more in terms of physical effort to write than to speak, some people find that web-based communication encourages them to express themselves more clearly and to put greater effort into understanding the other person. Explaining a problem also requires reflection on the problem by using the PC as a medium. Some people find it easier to express complex ideas, reflection on experiences, or feelings via the web forum, knowing that they have time to finish the thought before eliciting another person's reaction to what they have expressed. This means that the act of writing about one's experiences can itself solve problems. “For me, writing meant solving the problem - myself”, a teacher reported. The exchange of emails with a counsellor creates an automatic transcript of all sessions. Compared to face-to-face counselling, audio recording requires time-consuming manual transcription if a written record is desired. From the counsellor's point of view, the automatic transcription of text-based counselling via the web forum also allows greater transparency for purposes of supervision, professional consultations and reflections. Both parts have the opportunity to reflect and to learn. The asynchronous nature of counselling provides both client and counsellor with the opportunity to reflect on thoughts, feelings and other reactions to the other person's words. It is our experience that the amount of time we spend reflecting on a given statement by a client is significantly greater in the case of text exchanges than in the case of live verbal exchanges. Written examples can also be used as practical case examples in counsellor training. When working without the normal visual cues, both client and counsellor need to be especially aware of assumptions they might make about the other person. The opportunity to reflect on these assumptions can itself be helpful to the counselling process (Mulhauser 2008). Why do we offer online counselling via a web forum, not only by using e-mail? An open web forum is also an opportunity to build a knowledge base. Users and visitors can log into the teacher’s web forum anonymously and read communication processes between users and counsellors. This may in itself be enough for them to solve their own problems. This will give them an opportunity to reflect on problems by means of textual meta-positions or references. Learning by sharing experiences seems to be the most common way to use the web forum as a knowledge base.

7. SUMMING UP Our experiences suggest that the forum is good for knowledge building, including information and decision making. Counselling in respect of concrete implementation in personal relationships should be done in a faceto-face setting. That is what we call a dual or blended guidance system. The purpose of this study was to describe the kind of activity that is carried out when the Teachers’ Forum is used. Web-counsellor activities have been analysed. The theoretical foundation has contributed to design indicators that are useful in the questionnaire. The theory-based analysis has showed us that a questionnaire should be developed that is divided into three main categories: knowledge-building, social support and professional guidance. The site has also a potential to generate and apply knowledge. Knowledge construction as implemented can be used both by users and visitors. The large number of visitors shows that this site in practice also works as a database for a very large number of users. Users are fewer in number, but they use the site settings to return to specific questions and are thus in a dialogue with the site's supervisors. These are experienced teachers and those with experience as professional guides. Teachers often work alone with their problems. The site seems to have acquired a function both of social support and of professional guidance. For teachers it seems to be positive to be able to discuss their problems and to find solidarity with others. The attitudes that are valued most from professional advisers are the experience of being included, the experience of self-worth, confidence and empathy. The survey will be continued and a deeper examination made through indirect analysis of the visitors’ use of the Teachers’ Forum. It is now planned that students should maintain contact with the university and training institutions after leaving it to go into full practice. Teachers will have the opportunity to discuss their problems in a special forum and our students will have the chance to know the problems that a newcomer has to face. In this way

33

ISBN: 978-972-8924-82-9 © 2009 IADIS

we acquire important information about quality of teacher training, and our students will gain perspectives on practical problems and the motivation to learn about solving practical problems.

REFERENCES Argyris,C (1992) Bryt forsvarsrutinene. Hvordan lette organisasjonslæring. Oslo: Universitetsforlaget. Bråthen, I (ed) (1996) Vygotsky i pedagogikken. Oslo: Cappelen Akademiske. Caplan, G. (1970) The theory and Practice of Mental Health Consultation. New York: Basic Books. Curtis J. Bonk, Charles R. Graham (December 2005). The Handbook of Blended Learning: Global Perspectives, Local Designs (excerpt), Pfeiffer Wiley. ISBN 0787977580. Retrieved on 2006-12-26. Handal G. and Lauvås P. (1982) På egne vilkår. En strategi for veiledning av lærere. Oslo: Cappelens forlag. Imsen, G. (2000) Elevenes verden. Innføring i pedagogisk psykologi. Oslo: Tano Aschehoug. Krøger, H. & Reisky, A. (2004) Blended learning – Erfolgsfaktor Wissen. Bielefeldt: Bertelsmann. Lave, J., & Wenger, E. (1990). Situated Learning: Legitimate Periperal Participation. Cambridge, UK: Cambridge University Press. Lund, Ø. (2005) Nettbasert læring i spesialpedagogisk tjenesteyting. Tromsø: Det samfunnsvitenskapelige fakultet, Universitety of Tromsø. Mason, R. (2003): Sucessful online learning conferences: what is the magic formula? I Arneberg P. (ed): Læring i dialog på nettet. SOFF skriftserie 2003 no.1. McLellan, H. (1995). Situated Learning Perspectives. Englewood Cliffs, NJ: Educational Technology Publications. Mulhauser,G. (2008 ) Advantages of Therapy or Counselling by Email. http://counsellingresource.com/counselling-service/online-advantages.html Accessed January 25th 2009. Schön, D. (1983) The Reflective Practitioner. How Professionals think in Action”. New York: Basic Books Schön, D. (1987) Education the Reflective Practitioner. London: Jossey-Bass-Publishers. Sorensen, E.K. (2003) Designing for online dialogue and discussion in collaborative knowledge building networks. In Arneberg P. (ed): Læring i dialog på nettet. SOFF skriftserie 2003 no.1. Valiathan, P. (2002) Blended Learning Models, ASTD’s Source for E-learning. Learning Circuits. Virginia – USA. Vygotskij L./ Kozulin A. (ed 2001) Tenkning og tale. Oslo: Gyldendal Akademiske.

34


REFLECTIONS UPON ETHICAL CHALLENGES OF IT-PROFESSIONALS Darek M. Haftor Växjö University, Sweden

ABSTRACT This text presents research in progress, investigating normative and particularly moral considerations for the ITprofession. Recognizing that Information Technology may be both for good and for bad and that it has a unique capability to instrumentalise human reason, a structural investigation is presented of the normative questions inherent in the core tasks of the IT-professionals. While conventional ethical contributions, to the IT-profession, focuses mainly on the very usage of a deployed IT-artefact, proposing a set of rules, or a code of proper conduct, this inquiry switches the attention to the preceding process of IT-artefact development and management and to the type of moral dilemmas that ITprofessionals typically face – without providing a prescription for the solution. KEYWORDS Information and Communication Technology, ethics, morality, IT development process

1. INTRODUCTION This text reports some preliminary findings from a recently initiated inquiry into the normative considerations of the IT-profession. Information and communication Technologies (IT) are embedded in merely most of human and social affairs, and this embedding is most likely to continue. The assumption here is that as any technology, IT may be both for good and for bad, dependently on how it is utilised. Unlike any other technology, however, IT has a unique capability to affect our world, which requires careful normative and particularly moral considerations. While most of such moral inquiries today seem to address the question of how to use an ITartefact that is constructed and deployed, the present inquiry focuses on the moral conditions of the preceding process, that is the development process of the IT-artefact, and its master: the IT-professional. We start with an inquiry into some key moral characteristics of IT itself. Thereafter, an elaboration of the norms embedded into the process of the IT-artefact development and management is presented and is followed by an account of some of the key stakeholders involved in that process. This process and these stakeholders produce the structure of the here proposed conceptual framework for moral reflections of the ITprofessionals and of the IT-artefact development and management. Unlike other contributions to this area, we do not attempt to provide a code of moral conduct for the ITprofessionals, which must inevitably be based upon some choose ethical values. The here proposed framework aims rather to guide the moral reflection of the IT-professional independently of the moral convictions, e.g. utilitarian, deontological, or other, which is illustrated here. As the here presented conceptual framework is still in its initial stage and requires further elaboration, the text ends with some suggestions for its further development.

35

ISBN: 978-972-8924-82-9 © 2009 IADIS

2. KEY CHARACTERISTICS OF INFORMATION & COMMUNICATION TECHNOLOGY What are the peculiar characteristics of Information and communication Technologies (IT)?1 That is if we compare IT with, for example, nuclear-technology, bio-technology or any other type of technology. Instances of IT may include such as the book printing technology, radio, television, telephones, monitoring and controlling systems in nuclear power plants or surgery rooms, a bank’s automatic teller machine, the Internet – with its various offerings such as electronic journals, e-books and e-shopping – car dashboards with its information providing systems, video games, automatic customer payment systems in the supermarkets, this among thousands of other instances! It seems that IT is immersed everywhere in our human and social affairs, whether it is private, public, social, or professional. In total, all these systems, in one way or another, support and enable us to conduct certain activities. If we would ask our selves: Would I like to live in a situation where all this IT is not around…? We assume that most of us would say: No! Yet, there are both benefits and harms of any technology. All technology has the ability to instrumentalises human actions – that is to enable the conduct of human activities. However, the uniqueness of IT is constituted by its ability to instrumentalise a certain area of human action: human reason! Hence, while other technologies may instrumentalise human physical activities, such as digging a ditch, IT instrumentalises human mental2 activities, such as arithmetic calculations – this is a unique characteristic of IT; no other technology has this capability of instrumentaisation of human reason. This has profound implications in terms of IT’s capability to impact our world! The above mentioned examples of IT-systems do manifest the vast benefits that IT provides man, whether it is the elimination of dangerous acts for man, conduct of lifesaving activities, but also delivering pleasure and joy. On the other hand, IT may generate human harm. Stanley Kubrick’s seminal movie ‘2001: A Space Odyssey’ is an icon for the harm that instrumentalisation of human reason may lead to! The all controlling spaceship’s main computer, Hall 9000, is sufficiently intelligent to mimic the kind of unenlightened self-interest that is all too human! IT’s ability to instrumentalise human reason provides two generic abilities – automation and communication – that in turn exposes man for two generic threats. One relates to automation which enables the conduct of control, which in turn threats human integrity! The second relates to communication that enables mediated human communication that in turn threats the authenticity of human communication. Control may show to be helpful for man, for instance health monitoring systems that continuously monitors key health variables of a person and forwards them to a physician for a diagnosis. On the other hand, unwanted control of our shopping behavior, as conducted by some major supermarket chain, is what many of us would not want to be subjected to, as it violates our integrity! Likewise, IT mediated communication, like e-mailing, has enabled never before experienced communication opportunity between people and organizations. On the other hand, there are challenges, for example the ever increasing amount of the so-called SPAM, (unwanted commercial and other mailings, some estimate that more than 90% of all Internet traffic is SPAM) exposes us for false proposals, such as banking proposals that attempt to deceive us financially. Both the problem of control vs. integrity and of communication vs. authenticity is not generated by IT – it has been around for a long time. What is new is the IT context; it is the speed, the magnitude and the scope that the two problems assume: integrity and authenticity may now be violated in a way that we have never experienced! So far we have attempted to establish that IT is a unique technology as it has the ability to instrumentalise human reason, which in turn may generate benefits and harm to man and social affairs. This opens the quest 1

In broad terms, IT is assumed here to be an Information Processing System (after Newell & Simon, 1972), which stipulates that all IPS, whether natural or artificial, have the following four fundamental functions: to generate information, to transform information, to store information, and to transfer information. In this context we are interested of technological IPS only, hence Information and Communication Technologies, while disregarding form natural IPS:es, such as the neural systems or the genes. We assume that IT has two inherent generic benefits that may be understood as its generic purposes: one is to inform and the second is to automate (Zuboff, 1988). 2 This elaboration does not address the classical question of Artificial Intelligence of whether an information processing system can or can not duplicate human or natural intelligence. In this context, we simply hold that IT enables duplication of certain human mental activities, such as those symbol manipulating.

36


for normative reflections as to how we ought to utilize IT so that it creates benefits and not harm? In this regard we wish to articulate a further distinction that differentiates between IT that is implemented in some human affairs to serve some ends and utilized so or otherwise, on the one hand, and the very working process of IT development and implementation in human affairs, on the other hand – the latter is a precondition that generates the primer. Ethical reflections upon the utilization of IT, such as how to use Internet to create benefit and not harm is crucial for the wellbeing of human affairs, and a significant amount of reflections have been published. However ethical reflections, on how to develop IT, are also crucial as these open and close the normative and functional space of the IT utilization, and in that manner, these ethical reflections on IT development are more fundamental than those ethical reflections on how to utilize IT. Unfortunately, ethical reflections upon the development process of IT have generated limited amount of attention only; and where so, of a particularly type; this is the concern of the present elaboration and proposal. The existing typical reflections made are in terms of stating proposals for a code of ethics for IT professionals, such as the “ACM 3 4 Code of Ethics and Professional Conduct.” An example of its moral imperatives is: “Contribute to society and human well-being.” While we welcome this particular moral statement, such an approach to ethical meditation has some limitations. Firstly, it is very generic in its nature, meaning that it applies to everything, most of people would probably subscribe to its intention on intuitive bases, yet it does not say much about how to fulfill it, nor what it means to fulfill this imperative in a particular real-life situation. Secondly, and more importantly here, such imperatives provide a preference to a particular type of moral preferences and value sets – which is also needed (e.g. Johnson 2001). Yet, not everyone submits to a particular moral imperative or value set, for a variety of reasons. Therefore, the present elaboration focuses on a meta-level of such professional ethical norms, not subscribing to a particular ethical preference – whether it is utilitarianism or other type or consequentialism, some deontological position, or any type of virtue ethics. The focus here is rather to provide a structural guidance for how to conduct moral reflections and considerations for IT professionals in their profession. Therefore, in the next section, the structure of the IT development process is characterized in order to create a foundation for the forthcoming proposal for a framework to guide such reflections.

3. CHARACTERISATION OF THE IT-ARTEFACT LIFE-CYCLE A characterisation of the core tasks of the IT-professionals is now provided, functioning as a foundation for the forthcoming framework for ethical reflection of the IT-professional. The starting assumption here is that the central subject matter of an IT-professional is the IT-artefact 5. Therefore, the key question here is: what is done to this artefact? There are numerous so-called methodologies and processes for the development and management of IT-artefacts which seem to encompass similar tasks, even though organised in various manner. This inquiry assumes therefore one generic notion of such a process, including the following key phases: the inception, the analysis, the design, the construction and testing, the deployment, the maintenance, and the abolishment6 – see Figure 1 for an overview. In the following, we shall characterise briefly each phase of the IT-artefact development and management, as assumed here, and thereafter present some of the key ethical considerations inherent in these phases.

Inception

Analysis

Design

Construction

Deployment

Maintenance

Abolishment

Figure 1. Illustrates a process for the development and management of IT-artefacts, as assumed for the present elaboration.

3

Association of the Computing Machinery – see http://www.acm.org/about/code-of-ethics Other similar propositions may be found at IEEE: http://www.ieee.org/portal/cms_docs/about/CoE_poster.pdf 5 In very general terms, IT-artefact refers here to mans’ creation of information processing technology, where technology includes both physical objects and non physical human procedures and human organisation. 6 The mentioned order is logical, not necessarily temporal. 4

37

ISBN: 978-972-8924-82-9 © 2009 IADIS

3.1 The Inception The Inception is where the need or opportunity for a new (or changed) IT-artefact emerges. As we understand it, it is the least researched phase and at the same time one of the most important. The emergence of the idea for a new IT-artefact may take place in a formal organisational meeting but also when a customer complains or when an employee is on vacations. However when the mental consciousness of a need for an IT-artefact emerge, it is typically followed by some type of organisational formalisation, for example a pre-study with a set of resources allocated to its end. This pre-analysis, frequently constituted by an opportunity and feasibility analysis that may also include a business case diagnostic, will generate some kind of recommendation for the further steps: go or no go! It is important to recognise that this emergence of the awareness of a need or an opportunity takes place in a pre-given context. This pre-given context is constituted by existing determinative and normative structures that open and close the opportunities for further development. They may include the physical situation – people, buildings, factories, machines – the mental situation – previous experiences, education and training – cultural situation – differences and similarities in the opinions and suggestions and political structures involved – aesthetical preferences, legal stipulations, or ethical positions of what is right or not? In this phase, that characterises a transformation from current to an idea or proposal for an initiative, key moral questions that emerge include: who is to approve whether the mentioned pre-analysis is to be conducted or not, who is to be defined as responsible for the conduct of such a pre-analysis, and who should be involved in this pre-analysis; what kind of knowledge should be included in this pre-analysis and what kind of preferences (values) should be assumed (e.g. economic aiming at business benefit of the proposal, political aiming at power domination of the proposal as may be the case with the unions)? Another type of considerations regards the very content of the recommendation from the pre-study. Whether it is positive, i.e. to go ahead with the next phase, or it is negative, i.e. to stop; the decision is a normative standpoint on what ought to be or not!

3.2 The Analysis and Design The analysis work in the context of the development and management of IT-artefacts may contain different key activities with a variety of scopes, which may also overlap with the forthcoming phase, the design – hence the distinction between analysis and design varies however together they typically encompass an area that is commonly agreed upon. Therefore the distinction assumed here between analysis and design does not differ in total from the common notion of these two phases taken together. In short, the analysis work focuses here on the understanding of how the current situation is, in terms of human operations and actors, their processes and resources, and then the information involved, as well as the information processing instruments, such as the hardware and the software. The outcome of analysis, as understood here, is the detailed characterisation of the current situation. Next is the Design phase that aims to generate a design of the desired IT-artefact. While the work within the Analysis phase is concerned with the current situation the work within the Design phase regards the desired or ought to situation, in similar terms as the analysis.

3.3 The Construction and Deployment In the present conception, the Constructions phase of the forthcoming IT-artefact is conceived in terms of two sub-phases of work: the so-called system configuration or coding and then its various tests, such as logical tests and user tests. Each of these, in turn may be conceived in terms of two sub-stages: the planning of construction and the planning of tests, and then the conduct of construction and of tests. Given that the new or modified IT-artefact is designed and constructed it is about to be deployed within its desired context, such as an organisation, or a mobile-device. This requires again both planning work and then execution work, which may require various other specialised skills such as organisational change management, education and training.

38


3.4 The Maintenance and the Abolishment The Maintenance of an IT-artefact may involve a set of various activities, dependently on the actual situation. One central sub-phase is the evaluation of the IT-artefact in relation to some performance criteria, whether a priori defined or emerging during the evaluation process. A second central type of activities to be conducted are those that require that the current IT-artefact can maintain its operation and performance as it is – for example, a database typically requires some regular de-fragmentation in order to maintain its data access performance. Further, if software includes some errors then adjustments need to be made, which is a type of problem fixing. This sometimes leads to questions whether comprehensive adjustments should be classified as a development work that requires its own analysis, design, construction and deployment – which is clearly a normative judgement to be made and justified by someone. While few readings seem to address the very issue of the abolishment of an IT-artefact, it is a crucial and challenging work-stream! For example, a frequent challenge is what to do with the data in the database system? What to do with the physical artefacts after the closing down of the system? Who should decide whether the system should or should not be closed down?7

3.5 The Pattern of Moral Considerations within the IT-Artefact Life-Cycle In summary, we have presented here a structure of core tasks of the IT-professionals, organised into a set of phases, which are centred on the IT-artefact. As suggested previously, the presented structure is not said to be the only way to organise successful IT-artefact development and maintenance. Yet, the position here is that the central working tasks that should belong to any such a characterisation, are included here, which serves the key purpose of this presentation. The above characterisation of the IT-professionals’ core tasks distinguished between two kinds of normative questions. One addresses the foundational norms of each phase of tasks, such as who should be included in its conduct, what type of knowledge and values should be employed in its conduct; we shall refer to these as the predisposition norms, as in a way they predispose what can and cant not be achieved during the very conduct of the specific tasks of the IT-professional. The second type of norms addresses the very content of the outcome of each phase, such as what should be the content of a performed analysis or what should be the design of an IT-artefact – this type of norms are here called as terminal norms, as they are included in the very outcome of a set of professional tasks. To be sure, one and the same set of predisposition norms may give rise to variations of terminal norms. Also, there clearly is a relation, between the terminal norms generated in one phase, say analysis, and the predisposition norms assumed in the consequent phase, the design. For example, if the scope of an analysis is limited to a particular department of an organisation, then the design phase will most likely also be concerned with the selected department and not much more. However, this relation is only influential not casual as no logical derivations or deductions can justify in them selves what ought to be from that which is! For example, should the design of a new pay-desk or checkout in the supermarket imply that the cashiers will be fired and that older people will have to learn how to handle fully automated check-outs? If so, why and who is to decide this and on what grounds?

4. CHARACTERISATION OF THE KEY STAKEHOLDERS OF THE ITARTEFACT LIFE-CYCLE Given the above presented process for the IT-artefact development and maintenance the next central area of concern here is the kind of stakeholders typically involved in the above presented process of the ITprofessionals. These stakeholders contextualised within this process provide a certain space of moral considerations.

7

All the mentioned types of working tasks for the IT-professionals suggests that there may emerge subspecialisations, which is the typical cases, for example, System Analyst, System Designer, Database Manager, System Tester, Maintenance Engineer, etc. In the present elaboration, all these are regarded as subclasses of the general category of IT-professional.

39

ISBN: 978-972-8924-82-9 © 2009 IADIS

Starting with the IT-professional, and the Fellow-colleague IT-professional8, the interest of this stakeholder emerges typically from the assignment given: to develop and manage an IT-artefact, in relation with certain explicit given and some implicit norms. For example, the IT-professional may be given the assignment to develop an IT-artefact within a given financial budget and a limited time frame; or the ITprofessional is supposed to train and motivate the users of the IT-artefact so that they will actually use it! We may thus characterise this interest as related to the assignment in relation to the IT-artefact. The ITprofessional and its fellow-colleague give rise to a certain relationship of fellow-ship as they both have the same interest, however there may emerge yet another type of interest caused by an eventual dependency, when certain tasks of one IT-professional will determine the work of another IT-professional later on in the process of the IT-artefact development; for example the design work is dependent upon the analysis, while the maintenance work is dependent upon all the previous work conducted. This type of dependence creates a need for responsible actions and considerations between the IT-professionals. The assignment to develop an IT-artefact is typically given by what we wish to call here the Owner of the IT-artefact. The owner is someone, a person or an organisation, that has the power to initiate and terminate an IT-artefact initiative. Frequently, the owner may be a company that assigns another company to develop an IT-artefact. In this sense the owner has an interest of the IT-artefact, in terms of using it in a certain manner, such as automating certain business processes in its organisation or to sell it further on. The two above mentioned stakeholders, or roles, generate a central relationship, the IT-artefact owner and the ITProfessional relation, where the primer dominates the relation in terms of determining what kind of ITartefact to create and manage while the IT-professional determines the how to do this. A third central stakeholder is the Employer of the IT-Professional. Typically, the IT-professional is employed in a company and then assigned tasks or area of tasks. The interest of the Employer is thus that the IT-professional conducts his or her tasks with proper quality and in an efficient manner. In this sense there emerges a relationship between the IT-professional and her or his employer where the employer dominates the IT-Professional in terms of having the power of assigning him or her to a certain project and/or area of tasks, and also setting certain expectation on the quality and efficiency of the IT-professional’s work. On the other hand, the Employer is dependent upon the work of the IT-professional, in order to be able to generate the outcome committed, typically to the owner. A fourth central stakeholder is the (to be) User of the IT-artefact. It is typically a role, whether an organisational (e.g. a secretary using an invoice management system) or social (e.g. a friend connecting up with a friend on the Facebook). In this sense, the IT-professional both opens and closes opportunities for the user’s usage of the IT-artefact, which is a position with responsibility. An invoice management system may create a lot of challenges for a secretary if it requires new types of tasks to be conducted, and the opposite: it may eliminate part of the administrative burden of a secretary. In this sense, the IT-professional is in a position with a responsibility toward the user of the IT-artefact. We have now reviewed the stakeholders here assumed to play the key role for an IT-artefact development. This identification may never be assumed as final, and other type of stakeholders may be included where and when needed – for example the Public stakeholder or the User’s Client, or the Educator of the IT-professional. However, for the proposal put forward in the present elaboration, the listed stakeholders will suffice.

5. A GUIDE FOR THE NORMATIVE CONSIDERATIONS OF THE ITPROFESSIONAL In the discussion above, two structural sets were elaborated, the process of the IT-artefact development and management, as a space of tasks for the IT-professional, and then the key stakeholders for an IT-artefact development and management. These two sets may be unified into one two-dimensional framework generating a set of cells, as a result of the interaction between the two dimensions – see Table 1. Assume a situation in the construction phase when the IT-artefact is constructed and then is subjected to its various tests. The following normative questions may emerge: should the to-be-user be utilised as the 8 A sub-set of the IT-Professional that may be conceived here is the vendor / supplier category, which is assumed to be included in the former and more general category.

40


tester or not? Further, in the case where there are various segments of to-be-users, such as those that are well acquainted with IT-artefact usage vs. those that are not? Should both groups, in such a situation, be used as test-persons for the acceptance tests? In this manner, each phase will raise a set of normative consideration for each stakeholder involved, including their interrelations. This means that the here presented framework is not a prescription for what to do or not, rather a guidance for the reflections upon the various stakeholders’ interests in the various phases of the lifecycle of an IT-artefact. Table 1. Illustrates the here proposed framework for normative reflection of the IT-professionals. Inception

Analysis

Design

Construction

Deployment

Maintenance

Abolishment

IT-professional Fellow IT-Prof. Owner Employer User Other….

5.1 Normative Considerations Form a Particular Normative Standpoint The here proposed framework for the guidance of normative considerations of the IT-professionals is a guide for what type of reflections could be conducted; it intentionally is not a set of direct recommendations for how to decide in a particular professional situation. The actual decision for what to do or not to do is determined highly by the values assumed by the IT-professionals. In the following we provide an example of this by contrasting a situation and its decisions with two distinct ethical positions: the utilitarian9 approach and a deontological10 approach. We assume here the example of the need for test-groups of a new IT-artefact: a new inventory management system. One of the IT-professionals involved observes that the selected testers are younger women only and suggest to the chief tester to also include older people that may have less experience of ITsystems and more experience of inventory management. However, such an expansion of the testing may threaten the budget and time-delivery, which neither the employer nor the owner is of interest. The utilitarian position would investigate the positive and negative consequences, compare them and decide for the one with most positive consequences. Assume here then that the negative consequences of expanding the tests as suggested would be budget overrun and time-delivery overrun, which would in turn trigger a penalty for the IT-company that produces the IT-artefact, implying a decrease of payment with 10%. This in turn would imply that the IT-professionals would not be paid-out bonuses for a delivery on time and on budget. Further, this could also endanger the relationship between the IT-company and its client, for further business opportunities. Assume further that on the positive side there would only be the fact that a segment of the targeted users of the IT-artefact would be given the opportunity to evaluate the IT-system and provide input for possible adjustments to this segment of users. In such a case the utilitarian decision would clearly go for not expanding the tests. On the other hand, a deontological position, such as the Kantian, would make the opposite decision. This is so as its categorical imperative stipulates that no human being should be utilized merely as a means (or instrument) toward a higher goal; rather humans should be the end in themselves.

6. SUMMARY The introduction of contemporary information and communication technologies in human and social affairs opens opportunities and challenges. Unlike any other technology that human kind has conceived of, this technology can instrumentalise human rationality and thereby has a unique capability to impact our world!

9

The Utilitarian Approach we understand it as the moral imperative that everyone ought to act so as to bring about the greatest amount of happiness for the greatest number of people. 10 The Deontological Approach we understand it as the moral imperative to never treat another human being merely as a means but always as an end.

41

ISBN: 978-972-8924-82-9 © 2009 IADIS

This unique capability brings the need of careful responsibility and normative considerations both in the very use of this technology and during its very development – the latter is the concern of the present investigation. A conceptual framework is presented here to guide normative, particular moral, reflection that the ITprofessional needs to do during the development of an IT-artefacts. This framework has two dimensions, one that characterises the lifecycle of an IT-artefact while the other that articulates a set of key stakeholders of the artefact. In each phase of the lifecycle artefact, a set of normative questions may be defined followed by a search for answers. Unlike other normative guidelines, this proposal does not assume a particular moral standpoint in the specific questions it rather opens up the space of normative considerations to be made. In this sense, an IT-professional, of whatever moral conviction, may be guided in her or is reflections, within being imposed a set of actual values. The here proposed conceptual framework is still in its early phases, requiring further development. One is in the assessment whether the two dimensions of the framework make justice enough to the empirical situations of IT-artefact development and management? Secondly, a set of guiding questions for each phases of the lifecycle and each stakeholder, would facilitate and direct further the IT-professional’s reflections.

REFERENCES Bateson, G. (1972). Steps to an ecology of mind, Ballentine Book., New York. Franklin, U.M. (1999). Real World Technology. House of Anansi Press, Rev.ed. Johnson, D.G. (2004). Computer Ethics. 3rd ed. Prentice Hall, N.J. Newell, A., Simon, H.A. (1972). Human problem solving, Prentice-Hall, Inc. Englewood Cliffs. N.J. Shannon, C.E., Weaver, W. (1949). A Mathematical Theory of Communication. University of Illinois Press, Urbana Illinois, 11th ed., 1967. Simon, H.A. (1960). The New Science of Management Decision. New York, Harper & Row Pub. Zuboff, S. (1988). In the Age of the Smart Machine. The Future of Work and Power. Heineman, Oxford.

42


IT PROFESSIONALS’ PERCEPTIONS ON THE NATURE OF MORAL PROBLEMS Tero Vartiainen Turku School of Economics, Pori Unit P.O.Box 170, FI-28101 PORI, FINLAND

ABSTRACT This study reveals IT professionals’ perceptions on the concept of moral problem and what it is like to be in such a situation. Twenty-one IT professionals were interviewed and phenomenography was used in the analysis. The collective description shows that moral problems are perceived both as compulsory decision-making and compulsory wrong-doing situations. Moral problems are felt to be mentally hard but they involve a developmental opportunity. These results are reflected to the literature and implications for computing associations, firms and educational institutes are presented. KEYWORDS Moral problems, IT field, phenomenography.

1. INTRODUCTION Computer ethics literature recognized a variety of moral issues IT professionals confront. Such issues are privacy and databases, intellectual property rights, quality of software to name a few of them (e.g., Johnson 2001, Quinn 2006). To tackle moral problems, codes of ethics and conduct have been defined by professional organizations – see for a collection in Berleur and Brunnstein (1996), and computer ethics has been included in computing curricula (e.g., Gorgone et al. 2002). Albeit the importance of research and teaching of computer ethics is recognized there are few qualitative studied aiming to understand IT professionals’ experiences on morally challenging decision-making situations in the field. These situations have been called as moral dilemmas, moral conflicts and moral problems. Moral dilemma is perceived as a special case of moral conflict (Mason 1996, 216): moral conflict is perceived as resolvable but moral dilemma is perceived as irresolvable. In colloquial language, moral problem is used, and in this study for the sake of simplicity, it is conceived as synonymous to moral conflict. Like Joseph and Efron (1993) state about research on teacher’s moral problems that the dialogue should extend from ethical codes and curricula to consider a teacher as a moral agent who confronts and resolves moral problems in his or her daily life – in the same way we should take the viewpoint of an IT professional confronting moral problems in his or her professional practice. Knowledge about IT professionals’ experiences on moral problems is vital for developing means to prevent, confront and solve them. This study takes a step forward and determines IT-professionals’ conceptions of moral problems and how they feel to be in such a situation. With this information recommendations for research and practice are inferred. This paper is constructed as follows. In the second section, literature on moral problems is briefly introduced, and in the third section, research design for data collection is presented. In the fourth section, results are presented and they are discussed in the fifth section.

43

ISBN: 978-972-8924-82-9 © 2009 IADIS

2. MORAL PROBLEMS IN LITERATURE 2.1 Definitions of Moral Conflict In philosophical literature there is debate about the nature and existence of moral conflicts and dilemmas (e.g., Gowans 1987, Mason 1996). A moral dilemma is considered to be a special case of a moral conflict (Mason 1996, 216). In a moral dilemma, the moral agent has confronted an irresolvable conflict in his mind. These kinds of situations have been called as, e.g., genuine moral dilemmas (Hill 1996, 173; Nagel 1987) to be distinguished from apparent moral conflicts, which are solvable. A moral conflict, if it is a solvable one, is not a moral dilemma. The practical and theoretical difficulty, which emerges from these definitions, is how to draw the borderline between irresolvable and solvable moral conflicts. Statman (1995) approaches this difficulty by taking into account feelings relating to conflicts and dilemmas. He considers moral dilemmas as a subclass of moral conflicts. In the case of a moral dilemma, the moral agent is unable to produce a solution and feels anguish about it, and in the case of a moral conflict, the moral agent does not feel helplessness about the conflict.

2.2 From the Recognition of Moral Conflicts to Implementing what is Right Confronting moral conflicts means recognition of a moral conflict, making a decision, and implementing it. Four Component Model (FCM) (Rest 1984) describe four processes about moral behaviour and reacting to moral conflicts: moral sensitivity, judgment, motivation, and character. James Rest (1984) organized existing research on the psychology of morality into four processes: moral sensitivity, judgment, motivation, and character – according to which an individual may morally fail in her actions. The processes aim to answer the question: “What must we suppose happens psychologically in order for moral behavior to take place?” Moral sensitivity implies awareness of how our actions affect other people. It also involves being aware of alternative actions and how those actions affect other parties. For example, a teacher may not notice that he favors boys at the expense of girls, but when someone points out to him his actions, he may begin to observe his own behavior in a new light. Moral sensitivity is a key component in recognizing moral conflicts – it is possible that an individual does not observe that the decision-making situation has moral relevance. Moral judgment is about judging which courses of actions are the most justified. As moral judgment develops, a person's problem solving strategies become more directed towards others and more principled in nature. Kohlberg's (1981) six stages of moral development are based on the theory that people change their moral problem-solving strategies as they grow. People at higher stages can understand the principles they used when at the lower stages although they no longer prefer them. Moral motivation refers to prioritizing moral values above non-moral values. Here a moral agent asks “why be moral?” It is possible that a moral agent is aware of an ethical issue (moral sensitivity) and knows what should be done (moral judgment) but is not fully motivated to do what she considers the right thing to do. For example, if by lying one can profit economically, one is to select between an economical value (profit) and a moral value (honesty). If one chooses to lie, one has failed in moral motivation. Moral motivation, as well as moral character, connects knowledge to action. Moral character refers to the psychological strength to carry out a line of action. A person may be weak-willed and if others put enough pressure on him to act immorally, he may fail in this component. These processes develop on people as they mature but the development can also be supported by educational interventions.

2.3 Avoiding Moral Dilemmas As consequences of moral conflicts and dilemmas we aim to prevent them in future. Regarding conflicts and dilemmas there are bad feelings stemming from, for example, unfulfilled duties, and, accordingly, predicaments of moral dilemmas motivate us to change our lives and institutions to avoid dilemmas (Marcus 1987, 188). It can be claimed that the existence of moral dilemmas produce a second-order moral requirement to structure life so that conflicts between basic rules and principles were minimized (McConnell 1996, 36). Marcus (1996, 45) even claims that ”..it is wrong not to take due care to avoid such situations.” (Marcus 1996, 45).

44


In a moral dilemma two “oughts” (e.g., duty, obligation, moral requirement) compete with each other in the mind of a moral agent. Because one of the “oughts” will not be performed, it will leave a residue, that is to say, bad feeling upon a moral agent, remorse or guilt or duty to apologize or to make amends (McConnell 1996, 36). Regarding the bad feeling, there are two components: 1) the experiential component, the negative feeling, and 2) the cognitive component, a moral agent feels that she has done something wrong and takes responsibility of it (McConnell 1996, 38). In the case that a moral agent feels regret, she has negative feeling but she does not feel that she had done something wrong. Feeling regret of a situation may be appropriate although the moral agent was not causally responsible for bad consequences. But in the case of remorse, the both components apply. A procedure created for solving moral problems in computing (Kallman and Grillo 1996) includes a step, in which the user of the procedure is guided to deliberate how a similar moral problem could be prevented in the future. Codes of conduct, ethical codes and even laws can also be interpreted to have a function of preventing major dilemmas arising.

2.4 A Critique on the Studies on Moral Conflicts McDonald (1993) criticizes studies on moral conflicts/dilemmas. According to him moral dilemmas can be accompanied with negative, aversive, attitude, or they can be viewed positively as part of personal growth. Philosophers’ traditional way to use so-called thin examples of moral dilemmas support the negative and aversive way of perceiving moral dilemmas. As an example of thin example, a runaway trolley case, in which a driver has to choose between staying at the preset tract, thus first killing five persons, and the second tract, killing two persons. Use of thin examples make us perceive moral dilemmas as situations which come from nowhere and which put “completed”, well-trained and experienced moral agents to deliberate and choose between courses of action. The other way to perceive moral dilemmas, to recognize and accept moral history of each dilemma, that is to say, moral, social and political factors affecting to the present predicament, is to perceive moral dilemmas as catalysts for long-term moral transformation and nodes in an ongoing process of personal growth. Instead of perceiving moral agents to act upon dilemmas, the dilemmas act upon us and change us.

3. RESEARCH DESIGN AND DATA COLLECTION Information about IT-professionals’ conceptions of moral problems was gathered along with a research of moral conflicts perceived by parties of a project course in information systems (Vartiainen 2005). Clients of the project course were selected as subjects for this study, and the aim was to interview the representatives of the clients a) representing different hierarchical levels (managers, specialists), b) representing different ages, c) representing both men and women. 12 managers, 9 technical specialists were selected making totally 21 IT-professionals (of which 2 were from communications field). The aim was to avoid male biased interviewee selection by selecting also women to be interviewed. However, most of potential interviewees were male and as a consequence there were only five women (ca. 24%) among subjects. At the time of these interviews, 33.3% of subjects were in their twenties, 42.9% in their thirties, 14.3% in their forties, and 9.5% in their fifties. Subjects were provided with the following questions “Describe what a moral problem means” and “Describe how does it feel to be in a moral problem.” Phenomenography was used in the analysis of interview transcripts. The phenomenographical research method was developed to study human understanding of specific phenomena (Marton, 1992). The aim is to identify and describe qualitative variation in individuals’ experiences of their reality (Marton, 1986, 31). The phenomenographic researcher seeks to attain a holistic view of a selected phenomenon, which - according to Marton (1995, 178) - is possible because there are a limited number of qualitatively different ways of experiencing it. What is characteristic of phenomenography is that it aims at conceptualizations that are faithful to the individuals’ experience of the selected phenomenon. A phenomenographic researcher seeks qualitatively different ways of experiencing the phenomena regardless of whether the differences are between or within individuals. He or she tries to achieve a so-called second-order perspective on the investigated aspect of reality by describing the conceptions of a group of individuals - instead of taking the first-order approach and describing the reality directly, which is the convention in ethnographical studies. The outcome

45

ISBN: 978-972-8924-82-9 © 2009 IADIS

of a phenomenographic research project is the researcher’s interpretation of the subjects’ understanding of how the subjects experience, understand or perceive the studied phenomenon (Berglund, 2005, 37). The variation lies in the interpretation made by the researcher (Berglund 2005, 38). In the next section, the results of the phenomenographical analysis are presented.

4. RESULTS A collective description of the conceptions of moral problems follows. It consists of two classifications, the meaning of moral problems and the feelings related to them.

4.1 The Meaning of Moral Problems The subjects’ conceptions of moral problems are described below. There are two categories, compulsory decision-making and compulsory wrong-doing, which illustrate the compulsory element (Table 1). Table 1. Categories of meaning of moral problems Category Compulsory decision-making Compulsory wrong-doing

Description In a moral problem one is forced to make a decision. In a moral problem, one is forced to act in a morally reprehensible way (e.g., against a moral rule)

4.1.1 Compulsory Decision-making In this category, the subjects perceived that a moral problem involved being forced to make a decision, which could take the form of choosing from alternatives or just deciding what to do. The following extracts exemplify this category. Subject C7 was of the opinion that in a moral problem one was forced to decide on what was right in the situation: The researcher: “What comes to your mind about moral problems?” C7: “Well, you have to make a decision and there isn’t one single right thing there but you have to make a decision, you have to make a decision by yourself about what is right here. You know that it’s right from one viewpoint and that it’s right from another viewpoint as well. Subject C2 stated that a moral problem had two solutions, both of which were undesirable, and one should choose the lesser of the two evils: C2: “... there’s a problem, to which there are two solutions, but both of them are in a way poor. And then you should choose the lesser of the two evils but both of them are targeted, for example, towards an employee or towards anything…” Making a difficult matter known to someone may be a moral problem, and in cases like this one is forced to deliberate on how to express it. Subject C1 saw the way in which one expresses a difficult matter to one’s client as a potential moral problem: The researcher: “... what kind of perceptions do you have of what a moral problem is?” C1: “Well, of course it’s a very individual question. You may confront moral problems in working life, for example, in certain projects. If a project is late for some reason – how do you express it to the client? Here you might face a moral question…”

4.1.2 Compulsory Wrong-doing Here the subjects perceived that having a moral problem forced one to behave in a morally wrong way. That is to say that one is forced to break a moral or some other rule (of a firm or an employer), or forced to act against one’s own moral standards. The following extracts exemplify this category. Subject C1 said that many people might consider many necessary issues immoral:

46


C1: “... many people may consider issues that nevertheless are necessary immoral … to do or follow from the viewpoint of others or perhaps for the sake of the individual himself. But like I said people experience these issues in very diverse ways.” Later during the interview she continued by saying that sometimes it is not rational to conform to generally accepted ways of acting or to norms: C1: ... “you know what is the right way of living according to the norms. You might know the generally accepted way of dealing with issues, but still you might think that in a particular situation it isn’t rational or it may even be harmful. As a result we get these hard moral-ethical questions to resolve…” Subject C8 said that in a moral problem one felt that one had to do something that did not feel right: C8: “Perhaps it’s such [a situation] that you feel that you should do something that you know in your heart isn’t totally right. And that you have drifted into such a situation that you are asked to do something that is against your thinking, or that you have drifted into a situation from which the only way out is to do something that does not feel right for all of the parties” Subject C11 maintained that one was forced to act contrary to a moral rule: C11: “It could mean two things to me. The first is that I might have a personal moral problem about how I would behave in a certain situation because I might have to act contrary to a moral rule and according to another [moral rule], and they are in conflict with each other. Such situations often emerge in life. We talk about white lies…” I asked the subject to give an example of acting against a moral rule. She replied that when raising children or in traffic situations one was often forced to do so: C11: “Well, to give you a proper example. With children you’re forced to do this all the time. Lying, describing things so that they look fine, which is contrary to my moral rules – or, in the case of traffic, you may have to speed to be flexible.”

4.2 Feelings Relating to Moral Problems The subjects’ conceptions of how it feels to have a moral problem are described in the following. A moral problem was considered mentally hard, but also a character-development opportunity (Table 2). These categories express two distinct feelings. Table 2. Feelings relating to moral problems. Category A mentally hard situation

A development opportunity

Description A moral problem is a troublesome situation, in which one feels anxiety, depression, stress or other destructive feelings. A moral problem is challenging, and potentially character-developing, and after solving it one may have positive feelings such as pleasure

4.2.1 A Mentally Hard Situation The subjects conceived of a moral problem as mentally hard. They described it as a troublesome situation involving feelings of anxiety, depression, or stress, and said that it could give rise to conflicting and destructive feelings. The following extracts exemplify this category. Subject C5 thought that a moral problem was a destructive situation from which you could not escape: C5: “It’s a destructive situation – you can’t escape it. On the one hand you’d like to – on the other, you can’t. You meet similar situations in normal working life. As a supervisor you’re making decisions – decisions that should be brought to the notice of the personnel as soon as possible so that they can prepare for them. On the other hand, certain orders and rules may prohibit that.” Subject C6 viewed a moral problem as a very hard situation, which she could not tolerate for a very long time: C6: “It would surely have to be very hard. Personally, I wouldn’t have the strength to stay in such a situation for very long.” Subject C10 said that after making a choice one felt qualms:

47

ISBN: 978-972-8924-82-9 © 2009 IADIS

C10: “… there’s little harm as a consequence of a small problem but more with a bigger problem. It’s about how you have learned to handle these problems. Can you get rid of them or do they bother you for a long time? You get over the smaller problems with smaller qualms. In the case of a bigger problem, you may be forced to explain to yourself why you have done it in that way…” The researcher: “What are these qualms?” C10: “Qualms. A bad feeling. You know that you have done something that isn’t necessarily – that is to say qualms can be seen as a conflict with oneself. These issues are dependent on the person…” Subject C12 said that when one realizes that one has made a wrong decision one feels awkward: The researcher: “Yes. What else comes to your mind about how it feels to have a moral problem?” C12: “Well, well. When you have decided that you’re going to act in a certain way, and later you realize that ‘It is not okay! I have done wrong!’ then you have this – you have to start to handle it and to fetch the documents about what has been done, search for the emails and letters and so on. And when this happens, it’s an awkward situation.”

4.2.2 A Development Opportunity The subjects connected positive feelings to moral problems. They felt that such a problem was challenging and character-building, and that after solving it one might have positive feelings such as pleasure. Wrestling with moral problems forces one to deliberate on the situation carefully, which makes them challenging. In the long run, confronting them develops a person’s character. Solving a moral problem may give satisfaction and it may improve the atmosphere among people in a workplace, for example. The following extracts exemplify this category. Subject C12 considered moral problems challenging, and found discussions with colleagues very important: C12: “Such situations are challenges, and in my view, colleagues are very important in that you are able to discuss the problem and you are able to have a confidential discussion as well.” The researcher: “I see.” C12: “And you’re able to, well, I look for a sort of – in the case of harder situations – someone to discuss with and to test my ideas…” Subject C18 explained that a moral problem may be a depressing situation if one is not able to solve it, but solving it may even be pleasurable: The researcher: “Right. How does it feel to have a moral problem?” C18: “It might be depressing if you can’t do anything about it. But if you can solve it – well – let’s say that it’s possible that you might even feel pleasure in overcoming something…” Subject C21 considered a moral problem a detestable situation, but she also pointed out that confrontation with moral problems develops an individual. Moral problems and tackling them gives experience and wisdom: C21: “It is a conflict. On the one hand it gives you the ability to confront, in my opinion, it also develops stress-tolerance. On the other hand, it’s awkward to be in such a situation because you have to make a decision. Do you take it from your viewpoint or do you consider it from someone else’s viewpoint like a firm or an individual if it feels right?” The researcher: “Go on.” C21: “Well, there are some good and some bad points. When you confront them [moral problems], they give experience and wisdom – along with this experiential learning.” Subject C11 considered a moral problem a normal and even pleasurable situation, because one is forced to deliberate carefully: C11: “In my opinion, it’s a sort of normal situation, to which one is driven. I don’t consider it hard in any way. I know that all people confront them … – sometimes you’re forced to choose between hard alternatives. In my experience it can be, in a way, even pleasurable because you’re forced to deliberate and use your intuition. If everything were easy, what use would I have, then? I feel that working in this job and generally working with people, it’s a sort of salt of life when you’re forced to confront difficult situations that you’re able to resolve.”

48


5. DISCUSSION The collective descriptions of how IT professionals perceive moral problems and how it is like to be in such a situation show both of them bi-polarity: moral problems are compulsory decision-making situations in which one is forced to do something wrong and moral problems are mentally hard situations but they also bear developmental opportunity. The result supports the definitions on moral problems, moral conflicts and moral dilemmas in philosophical and moral psychological literature which states that in a moral problem one is forced to choose between competing moral requirements (Hill 1996; Nagel 1987) and that moral problems are mentally hard situations and that there are a variety negative feelings associated with them (Stattman 1995). The results also give support to McDonald’s (1993) critique on studies on moral conflicts: The subjects of this study perceived moral problems as developmental opportunities. The challenge which becomes visible from these results and the previous studies is that how to consider the paradox of human life: inevitable confrontations with moral problems force people to make decisions and commit acts which they perceive morally wrong. The situations are mentally hard but solving them challenge and develop people. Next, recommendations which aim to support IT professionals to tackle with moral problems are presented. Educational institutes should teach professional ethics. Ethics teaching should be integrated in computing curricula either by arranging ethics courses or by integrating ethics content to other courses. Such content should include theories of ethics and practical exercises on moral decision-making and moral argumentation with real-life moral conflicts. Recognition of moral history when using real-life cases. Moral problems bear developmental opportunity. Therefore, in-depth descriptions of real-life moral problems should take into account moral history (McDonald 1993) of the cases, ie., the social and political contexts in which the conflicts emerge. Professional associations might publish these stories in professional magazines, for example, and educational institutes could use these stories in ethics teaching, as a content for dilemma discussions (Boss 1994), for example. By contextualization of the moral conflicts and by becoming aware of the complex factors affecting to situations and integrating critical discussion on how to solve such cases would develop moral sensitivity and moral judgment (Rest 1984). IT firms should develop a business code of integrity. Corporations should develop a business code of integrity (Kaptein and Wempe 2002, 271) concerning the relationship between them and their stakeholders. Such a code should include virtue statement, principles statement, and mission statement. The virtue statement offers employers the flexibility to handle the moral problems they face, while the principles statement is more concrete and coercive. The mission statement should set the prioritizations otherwise it would create problems rather than prevent them. Kaptein and Wempe state that the power of codes most probably lies in the thinking process about what they should cover: it is a question of personal and collective reflection on the moral problems faced in the corporation, the responsibilities there are and the virtues that should be adhered to. This kind of discussion process and the code of integrity could produce frameworks for drawing up more concrete contracts with stakeholders. Professionals associations should support its members who confront moral problems. Pertaining to the mentally hard situation of confronting hard moral problems or moral dilemmas professional associations should offer help for professionals in such situations. Associations might form ethics committees or give advice in professional magazines, for example. They could also encourage and support IT firms to adopt business codes (see the above recommendation). In this way moral argumentation and discussion could be integrated as a natural phenomenon in the whole of IT field from the educational institutes to private firms. A research implication. For computing education means to develop students’ moral behavior in accordance with FCM (Rest 1984), for example, should be developed and the effects of educational interventions should be studied. In this way, effective means to support individuals in moral maturing may be found. Evaluation of the study. Although the interviewees represent one nationality and the majority of the subjects were male and the population was small the similarities between the results and the previous studies strengthen the validity of the results. As far as Lacity and Janson’s (1994) criterion is concerned, according to which the validity of interpretive studies rests on acceptance by the scientific community, it is up to the reader to decide if the results are meaningful in other contexts. However, the author has cited verbatim from he interview extracts to provide evidence.

49

ISBN: 978-972-8924-82-9 © 2009 IADIS

REFERENCES Berglund, A. 2005, Learning computer systems in a distributed project course. The what, why, how and where. ACTA UNIVERSITATIS UPSALIENSIS, Uppsala Dissertations from the Faculty of Science and Technology, 62. Berleur, J., Brunnstein, K. (eds.) 1996. Ethics of Computing, Codes, spaces for discussion and law. London: Chapman & Hall. Boss, J.A. 1994. The effect of community service work on the moral development of college ethics students. Journal of Moral Education 23, 183-198. McConnell, T. 1987. Moral Dilemmas and Consistency in Ethics. In C.W. Gowans (Ed.) Moral Dilemmas. New York: Oxford University Press. 154-173. McDonald, J. 1993. Thin examples of moral dilemmas. Social Theory & Practice. Summer, Vol. 19, Issue 2. 225-238. Gorgone, J. T., Davis G. B., Valacich, J. S., Topi, H., Feinstein, D. L. & Longenecker, H. E. Jr. 2002. IS 2002: model curriculum and guidelines for undergraduate degree programs in information systems. Communications of the AIS, 11 (Article 1). Gowans, C.W. 1987. The Debate on Moral Dilemmas In C.W. Gowans (Ed.) Moral Dilemmas. New York: Oxford University Press. 3-33. Hill, T.E. 1996. Moral Dilemmas, Gaps, and Residues: A Kantian Perspective. In H.E. Mason (Ed.) Moral Dilemmas and Moral Theory. New York: Oxford University Press. 167-198. Johnson D.G. 2001. Computer Ethics. Upper Saddle River (NJ): Prentice Hall. Joseph, P., Efron S. 1993. Moral Choices/Moral Conflicts: Teachers’ Self-Perceptions, Journal of Moral Education (22)3, pp 201-221. Kaptein, M., Wempe, J. 2002, The Balanced Company, A Theory of Corporate Integrity. Oxford: Oxford University Press. Kohlberg, L. 1981. The Philosophy of Moral Development, Moral Stages and the Idea of Justice. San Francisco: Harper & Row. Lacity M.C., and Janson M.A., 1994: Understanding Qualitatitve Data: A Framework of Text Analysis Methods, Journal of Management Information Systems, Fall 1994, Vol. 11, No. 2, pp. 137-155. Marcus, R.B. 1987. Moral Dilemmas and Consistency. In C.W. Gowans (Ed.) Moral Dilemmas. New York: Oxford University Press. 188-204. Martin, C.D. & Huff, C.W. 1997. “A Conceptual and Pedagogical Framework for Teaching Ethics and Social Impact in Computer Science”, Proceedings of 27th Annual Frontiers in Education Conference: Teaching and Learning in an Era of Change. Vol. 1. Marton, F. 1986, “Phenomenography – a research approach to investigating different understandings of reality.” Journal of Thought. 21 (3), 28-49. Marton, F. 1992, ‘Phenomenography and “the art of teaching all things to men”’. Qualitative Studies in Education 3 (3), pp. 253-267. Marton, F. 1995, “Cognosco ergo sum, Reflections on reflections.” Nordisk Pedagogik 15 (3), pp. 165-180. Mason H.E. 1996: ”Responsibilities and Principles: Reflections on the Sources of Moral Dilemmas”, Moral Dilemmas and Moral Theory (ed. Mason H.E.) Oxford University Press, New York. Nagel, T. 1987. The Fragmentation of Value. In C.W. Gowans (Ed.) Moral Dilemmas. New York: Oxford University Press. 174-187. Quinn, M.J. 2006. Ethics for the Information Age. Boston: Addison Wesley. Rest, J. 1984. The Major Components of Morality. In W.M. Kurtines, J.L. Gewirtz (Eds.) Morality, Moral Behavior, and Moral Development. New York: A Wiley-Interscience Publication. 24-38. Statman, D. 1995. Moral Dilemmas. Amsterdam: Rodobi. Vartiainen T. 2005. Moral Conflicts in a Project Course in Information Systems Education. Dissertation thesis. Department of Computer Science and Information Systems. University of Jyväskylä.

50


DIGITAL IMMIGRANTS, DIGITAL NATIVES AND CONTINUING PROFESSIONAL DEVELOPMENT (CPD) FOR ACADEMICS: ARE THE STEREOTYPES UNHELPFUL? Dr. Andrew Rothwell Centre for the Study of Higher Education, Coventry University

ABSTRACT ‘Digital natives’ and ‘digital immigrants’ were identified by Marc Prensky (2001a, 2001b) and describe in the first case young people of undergraduate age or less, who have grown up in rich, technologically enhanced environments and whose brains (he suggested): ‘are likely to be physically different as a result of the digital input they received when growing up’ (2001b, page 3). Those of us who may consider ourselves ‘digital immigrants’ may not have been ‘born into the digital world’ but we do embrace it, even if we may sometimes find it difficult. Much of the literature on these ‘immigrants and natives’ focuses on the student experience, the challenges of teaching and learning with the digital native generation. Examples include Zevenbergen (2007) on pre-school learners, Green and Hannon (2007) on school age learners, and Sharpe et al. (2006) and Salaway et al. (2006) on University-level learners. This paper takes a different perspective and reports research conducted with academics in the UK that seeks to redress the image of (many nontechnical) academic faculty as reluctant digital immigrants: rather it suggest that academic faculty, through exposure to their own technologically rich environment, and driven by the acquired ability to multi-task to cope with the multifaceted nature of their roles and the varied pressures of their jobs, have become ‘wired’ more like digital natives, and undertake their continuing professional development (CPD) in a more ‘native’ than ‘immigrant’ way, a factor which institutions ignore at their peril. At a time of increasing knowledge obsolescence (Pazy, 1992) faculty themselves may not be able to optimise their learning experiences because the media through which essential professional development is delivered has not kept pace with the changing demands (and approach to learning) of the intended audience. This in turn has consequences for the sustainability of the University in an increasingly competitive and challenging environment. Although the survey used has only been implemented in one institution so far, qualitative and exploratory work undertaken elsewhere (Rothwell et al. 2008), suggests that these issues may be common across a range of institutions, thus presenting implications for strategic approaches to CPD in many Universities. KEYWORDS Continuing Professional Development, Digital Immigrants, Digital Natives

1. INTRODUCTION Universities in the UK, in common with Universities throughout the developed world, are facing unprecedented challenges. They find themselves in the context of challenging domestic markets, the need to explore and develop new areas of activity, tough competition in a global marketplace, rapid technological advances, changing pedagogic practices, and significant financial challenges for some institutions. Academic staff must operate within the context of these challenges, and at the same time address the fact that disciplinary knowledge obsolescence is also a live issue for many contemporary professionals. At the University that was the focus of this study, in common with many others, there is a range of CPD provided, from short courses, qualification based programs, informal networking, formal staff development events, to research and scholarly activity. It is a mid-sized university, organized in four faculties (Engineering & Computing; Business, Environment & Society; Art & Design; Health & life Sciences; with some smaller units such as a School of Lifelong Learning, and the Educational Development Unit) located in the West Midlands region of the UK. Motivations for conducting the research included a perception (based on informal reports from a number of universities) of concerns about low attendance at, or cancellation of,

51

ISBN: 978-972-8924-82-9 © 2009 IADIS

formal ‘courses’ or staff development events; a view that there might be a potential mismatch between CPD needs with the delivery of CPD and CPD provision; and a perception that this might have something to do with the personal information-management behavior displayed by academic faculty. The paper aims to address the following research questions: • What are the attitudes and preferences of a sample of Academic Faculty in relation to their Continuing Professional Development? • Do academic faculty engage in CPD like ‘digital immigrants’ or ‘digital natives’? Does this (and related) terminology help us to understand the CPD behavior of contemporary academics, or does it do academic faculty a disservice? • Does the CPD provision typically offered in a contemporary University accurately reflect preferred means of engaging in CPD? If not, what adjustments need to be made? What are the implications for individuals developing digital CPD resources?

2. LITERATURE REVIEW: CONTINUING PROFESSIONAL DEVELOPMENT, IMMIGRANTS AND NATIVES. This paper presents a perspective on CPD from the point of view of academic colleagues who undertake it: what CPD they say they do, their attitudes to CPD, and what they see as CPD priorities both for themselves and for the University. CPD presents academics with both challenges and opportunities. One challenge lies in finding the time to keep up to date because academics serve in two dimensions: as subject specialists, and as teachers in our discipline context (Jackson 2005), and as a consequence have a range of drivers, internal and external, that require us to undertake CPD. As was found in this study, one of the features of University life is the diversity of roles and cultures contained within the academic commmunity. Some previous research has already been undertaken on academic CPD, and what individuals’ prefererences include. Some examples are CPD arising out of everyday work (Dunne 2005), from informal activities (King 2004), from research training (Chivers 2005), or CPD from curriculum design actvities (Gannon-Leary and McCarthy 2007). Generally speaking workshops and courses did not feature highly on the lists of priorities identified by these authors, yet these remain the main channel of staff development provision in most institutuions. There has been some research undertaken internationally (eg. Alexandrou et al. (2007), but this has tended to focus on national structures and standards. The UK’s Higher Education Academy (HEA) recognised the need to promote planned approaches to CPD, and supported (2006/7) pilot programmes in eight Universities, aligned to their professional standards framework. This model generally aimed to integrate career development, professional development, and the strategic development of Human Resources. HEA also found that informal methods of updating were preferred (as cited in Norris 2007). Norris quoted (page 31): ‘ CPD must focus on the personal and professional development of the individual in ways that are appropriate to their needs and working context. CPD ought not to be defined in terms of courses and events but should reflect a much more open understanding of ways in which individuals, or as communities of practice, may seek to develop and improve their professional practice’. In the context of this study, this literature (and the HEA Guidelines) suggested that there might be a disjoint between the professional development provided in Universities, and faculty preferences. The HEA’s promotion of formal approaches to CPD has also been reflected in increasing pressures at the University level to have strategic approaches to CPD, CPD policies, or even to require staff to provide evidence of CPD undertaken (Cope et al., 2008). While such perspectives may seem overtly managerialist, they do add to the pressure on individual faculty members. So how do academics relate to their working context, and learn within it and from it? Are our assumptions about academics in this respect valid, or dated? A potentially helpful explanatory framework was suggested by Prensky’s (2001a, 2001b) stereotypes of ‘digital natives’ and ‘digital immigrants’, typically defined (for a contemporary summary see eg. Toledo, 2007) as those who were or were not born into the digital world, or whose information literacy skills were defined in the digital or the print world (respectively) . Digital natives therefore represent the post-millenial schools generation, who are now presenting themselves as University-age learners. A plethora of literature of varying degrees of gravitas describes how they multi-task, interact, and assimilate information differently, building on Prensky’s suggestion that lifelong exposure to the digital world

52


has caused their brains to be wired differently. Accepting that ‘immigrants and natives’ represent sterotypes that are some distance apart, Toledo (2007, p. 87 et seq) summarized what she referred to as ‘embellishments’ to Prenskys’ original typology including digital recluses and refugees (use of computers not by choice); and the self-explanatory digital explorers, digital innovators and digital addicts, the latter as observed in the wi-fi lounge of an Austrian Ski hotel on Christmas Day by this author. Toledo proposes (page 88 et seq.) a further category of her own, that of digital tourist – ‘visitors in the digital world’. However one might choose to categorize individuals, both students and tutors now have the opportunity to create and source digital materials, to manipulate digital resources and engage with media-enhanced environments in a flexible way, and this has the potential to impact not just on teaching and learning, but also professional development. To sumarise and contextualize the literature reviewed, academic faculty face a range of pressures at the present time. These include the need to keep up to date in their subject discipline, exhortations both from their Universities and often external professional bodies to engage in (and evidence) CPD, keeping up to date with pedagogical developments including advances in learning technology, and, if the ‘immigrants and natives’ stereotypes are correct, coping with fundamental shift in the learning styles of their students. What appears to have been hitherto neglected in this field of research is an analysis of the strategies faculty employ to keep up to date in this context and a consideration of whether university provision is appropriate. The next section describes the research methodology adopted.

3. METHODOLOGY To explore the range of issues that would need to be considered in undertaking a research project across a diverse body of academic faculty, exploratory discussions were carried out in a range of fora, including: - focus groups with Teaching Development Fellows from across the institution, - faculty Teaching and Learning Committees (TALCs), - key informant interviews eg. HR colleagues, Deans and Assistant Deans These discussions, when linked with information from a literature review, and previous research on CPD for professionals undertaken by the author (eg. Rothwell & Herbert 2007), provided the framework for a survey design including questions for the survey instrument, definitions eg. staff categories, and the identification of some faculty-specific issues that would need to be addressed, for example, the importance of current professional practice in Art & Design, or of ‘communities of practice’ in Health & Life Sciences. The questionnaire design incorporated a range of demographic factors that were believed to be significant including gender, age, faculty, length of service, highest qualification, and whether mandatory CPD was undertaken for an external professional body. The next twenty-three items were a list of CPD strategies intended to elicit respondent preferences, derived from the exploratory research and also previous studies of CPD with other professional groups (Rothwell and Arnold, 2005; Rothwell and Herbert, 2007). Example items included ‘spontaneous learning arising from work or personal activities’, ‘exchanging emails on professional topics with other people’ and ‘University internal training courses’. Scoring was on a five point scale: frequently - quite often - occasionally – rarely – never. As the list could never be exhaustive, two ‘other’ items were added were respondents could include their own choices, but very few did. Seven items (selected and adapted from a nine item scale derived from Sadler-Smith and Badger, 1998) were a scale to explore attitudes to the perceived value of CPD. The internal reliability score was good, in this study, α = .84, N = 118 (c. 20% response rate). Scoring was on a five-point scale: strongly agree – agree – neither agree not disagree – disagree - strongly disagree. Example items included ‘Engaging in my own CPD can enhance my employability and career prospects’, and ‘engaging in CPD activity motivates me’. Finally, a number of individual questions related to perceptions of the development review process (appraisal), individual career planning intentions, personal development priorities, and what respondents perceived as being institutional development priorities. Although much of the survey instrument had been used in previous studies it was pilot-tested with a number of key informants and as a result some changes made. To minimize position bias in responses, two differentlyordered versions of the questionnaire were created. Administration was by email with endorsement from faculty Deans. Respondents were assured that data would be used only for research purposes, and of the anonymity of their responses.

53

ISBN: 978-972-8924-82-9 © 2009 IADIS

4. RESULTS One hundred and eighteen useable responses were received, 71 male and 47 female (60.2%/39.8%), and can briefly be summarized as follows: Sixty per cent of respondents were 46 or over, more in Art & Design (AD), Engineering & Computing (EC). As one might expect this was a well qualified sample with just 13.5% lacking Masters or Doctoral level qualifications. Almost 80% of respondents were full time, with Engineering the highest proportion, while nearly 80% of respondents were PL/Reader level (thus in terms of UK job classifications slightly more senior). Almost half of the respondents had 20 or more years experience in their field, and 63.6% of Health and Life Sciences respondents undertook mandatory CPD in relation to a professional body other than the HEA. Table 1. CPD Engagement, Preferences and Ranking Overall and by Faculty (top 5/bottom 1 items), full results available on request OVERALL 1

Reading of journals and books relevant to my subject

2

Acquiring knowledge through browsing websites/ surfing the net Designing a course or module

3

4

Spontaneous learning arising from work or personal activities

5

Subject based research for publication(s)

HEALTH AND LIFE SCIENCES

BUSINESS ENVIRONMENT AND SOCIETY Reading of journals Reading of journals and books relevant and books relevant to my subject to my subject

ART & DESIGN

ENGINEERING & COMPUTING

Reading of journals Reading of journals and books relevant and books relevant to to my subject my subject

Designing a course Designing a course Acquiring or module knowledge through or module browsing websites/ surfing the net

Acquiring knowledge through browsing websites/ surfing the net

Keeping a portfolio record of CPD activities I have undertaken Feedback from students

Reflective discussions with colleagues that are informal but still relevant to what I do Designing a course or module

Spontaneous learning arising from work or personal activities

23 Working towards a Working towards a qualification qualification which which is paid for is paid for by the by the university university

Acquiring knowledge through browsing websites/ surfing the net Reflective discussions with colleagues that are informal but still relevant to what I do Spontaneous learning arising from work or personal activities

Subject based research for publication(s)

Keeping a portfolio record of CPD activities I have undertaken

Keeping a portfolio record of CPD activities I have undertaken

Spontaneous learning arising from work or personal activities Working as a practitioner in my specialist field

“OTHER” Exchanging emails on professional topics with other people Working as a practitioner in my specialist field

Reading of journals and books relevant to my subject Reflective discussions with colleagues that are informal but still relevant to what I do Subject based research Spontaneous for publication(s) learning arising from work or personal activities

Working towards a qualification which is paid for by the university

Membership of committees at my place of work eg quality, health and safety

Table 1 presents extracts from the ranked lists of mean scores for CPD preferences, by faculty (department). Note that a higher proportion of Health and Life Sciences (HLS) respondents tended to slightly skew the overall results. Small numbers of respondents in the 'other' category (smaller sub-units within the institution) and Art & Design have tended to produce 'different' distributions with groups of responses on equal scores. Comments on the results overall include that there are some striking similarities in high scores: and journals (top ranked for all except 'other'), - Browsing websites (Top 3 for HLS, BES, EC), - Designing a course or module (Top 4 for BES, AD, EC), - Spontaneous learning arising from work or personal activities (top 6, all), - Subject based research for publication (never lower than 8th),

54


The first item here should not necessarily be seen as an endorsement of paper sources: online resources such as ‘Emerald’, ‘EBSCO’, ‘Psychlit’ and other journal resources (or even Google Scholar) have transformed the ability to search and retrieve information. There are some interesting low scores for items that one might have hoped to see score more highly, and which form the backbone of University CPD provision, such as ‘working towards a qualification that is paid for by the University’ (never higher than 18th), or ‘University internal training courses’ (never higher than 13/23 and 17/23 overall). There were also some notable differences, emphasizing differences in practice between the faculties, such as ‘keeping a portfolio record’, which ranked very low in all faculties except Health and Life Sciences, who are most likely to be required to undertake mandatory CPD. Overall these results reinforce the perception that there is considerable diversity in CPD practice and preferences between the faculties, arising from subject/disciplinary differences and in some cases linked to specific practices in the discipline based professions and that a University 'CPD framework' must account for this. It is reasonable to conclude, based on this sample, that: - Most academics learn in an unstructured, ad hoc way, juggling conflicting demands on their time, - Most are apparently not very interested in keeping portfolios, except where their discipline/subject or an external profession demand it, - Academics may not necessarily see the formal, structured training and development provision provided by the University as relevant, accessible, or even enhancing their CPD. In respect of the attitude scale for ‘perceived value of CPD’ (CPDV), the highest mean scores were recorded for respondents from Health & Life Sciences (M = 29.21, SD = 4.17) and Business Environment & Society (M = 28.87, SD = 3.24), the lowest were males in Engineering and Computing (M = 25.11, SD = 6.55) and females in Art & Design (M = 24.29, SD = 9.66). Scores were negatively skewed ie, on the higher side, as the highest possible score was only 35, so the mean of responses from most respondents, notably HLS was high. This suggests a positive attitude, on the whole, for these respondents, to CPD. One-way between groups ANOVA were conducted to explore the impact of a range of demographic factors on total perceived value of CPD (ie., were there variations by any grouping on how positively orientated respondents were towards CPD). No statistically significant differences were found in analyses by age, professional experience, gender (although women scored very slightly lower which is contradictory to most published research on CPD), qualification level, mode of work, job role (title), whether or not respondents were HEA members and whether or not they had to do mandatory CPD for another professional body. There was a modestly statistically significant difference in total CPDV scores between respondents in Health & Life Sciences, and Engineering & Computing [F (4, 112) = 2.12, p = .072]. The effect size, using eta squared, was 0.07, which is moderate. Post-hoc comparisons using the Tukey HSD test indicated that the mean score for the youngest (26-35) age group in Health & Life Sciences (M = 29.20, SD = 4.16), was significantly different from the oldest age group (over 56) in Engineering & Computing (M = 25.8, SD = 5.73). Responses to the questions about perceived individual and institutional priorities did not on the whole prioritize training in technological development (although computer-related updating did generally present itself in the upper quartile across most faculties), rather research, research skills and subject knowledge were seen as both individual and university priorities, reflecting on the one hand an institutional driver to become more research-led, and on the other concerns about rapidly advancing knowledge within the subject and disciplinary fields. Another noticeable attribute of the ‘needs’ items was a long tail of very individual concerns, again reinforcing the perception that ‘courses’ did not present the right sort of learning opportunities for this group of respondents. Finally, responses to the ‘other comments’ question emphasized more than anything the pressures on individuals in terms of workload: they did want to ‘do CPD’, but it was difficult, and they didn’t always feel supported in their endeavors.

5. DISCUSSION In terms of Continuing Professional Development, the study presented some encouraging news, and some issues for concern. On the positive side, respondents to the study were on the whole positive about CPD. They had embraced the University’s drive to promote research (in what had historically been a teaching-orientated institution); appeared keen to embrace new pedagogical developments, and to keep up to date in their subject areas. Those individuals who were required to undertake CPD for an external professional body also appeared

55

ISBN: 978-972-8924-82-9 © 2009 IADIS

keen to do so, and where appropriate to document their development in a portfolio of evidence. Respondents were also creative in the updating strategies employed, although this creativity was often a response to workload pressures. In this respect results were consistent with literature findings (eg. Dunne 2005, King 2004, Norris 2007). The University is advanced in its’ applications of technology enhanced learning, including the use of virtual environments, blogging by staff, digital repositories as well as the more usual online resources and respondents did not flag these up as a particular development issue. Data suggest that the CPD behaviour of respondents to this study is more akin to that of digital natives than digital immigrants in the modes of learning and preferences described: an emphasis on the informal and everyday; juggling different resources such as learning from email conversations and web-based sources, personal and professional networks, student feedback, and colleague interactions. Perhaps concerns about the invasion of Higher Education by the new species of digital natives will hold fewer concerns for these respondents that had been feared, if academics do buck their stereotype, and are more ‘native’ than ‘immigrant’ – or are certainly no worse than diligent digital ‘tourists’. On the other hand, the relatively low response rate must present cause for concern. While c. 20% is not uncommon in survey research, in this context it must be admitted as a weakness of the study. One former colleague described much educational development activity in UK universities as ‘talking to ourselves’. Where courses do run, they are often attended by people who are already well engaged with pedagogical developments, whether related to E-Learning or reflective practice, inclusion, retention or internationalization. This ‘extended family’ of educational developers is observable in many academic communities, and for those engaged in CPD, it is our ability to evangelize beyond this boundary, to reach the resolute Luddite poring over his (or her) dusty files that will be the measure of our success. Thus, while this study has suggested that the conception of ‘digital immigrants’ may do many wellengaged and highly professional colleagues a disservice, there remains a potentially hidden issue of lack of engagement with updating for a significant proportion of academics, about whose CPD behaviour little is known. This suggests a need for substantial further research into academics’ CPD behaviour, needs and preferences, both within the UK and internationally. This needs to be research that is significant not just in terms of teaching the ‘digital natives’ as a new generation of learners, but also the sustainability of individual careers, cumulatively impacting on organizational competitiveness. A deeper concern lies in those individuals who may not yet even be ‘digital immigrants’, potentially among the non-respondents to this study, and who risk falling further and further behind even a minimum standard for operational competence in day to day activity. The challenge lies in achieving a balance between imposing mandatory training (but it might come to that!) and engaging, voluntarily, the reluctant ‘digital recluses and refugees’ (Toledo 2007).

6. CONCLUSION The results reveal, for this sample, a potential disjoint between CPD preferences and CPD provision. Many academic staff have the facility, often born out of necessity, to juggle a range of conflicting demands on their time, have multiple roles, utilise a range of resources, are digitally-connected and technologically literate, and seek recognition for the informal and spontaneous learning that they engage in all the time, every day. Like digital natives (even if they aren’t ‘wired’ quite the same way) they too inhabit rich, digitally enhanced environments, and see no reason why their professional development provision can’t be dynamic, interactive, and capable of being captured and recorded instantly. Universities who fail to understand this are guilty of more than doing their faculty a disservice: they are doubly failing in promoting provision delivered by outdated modes of learning that does little to address the needs of those with the greatest learning gap. Finally, while there appears to be an assumption that faculty are in some way lagging behind in their use of technology, it is suggested that this may not in fact be the case due to their flexibility, adaptability and ability to multi-task, and that it is the presentation of CPD opportunities that is deficient. While academic faculty can never, by definition, be ‘digital natives’, they can be highly effective ‘immigrants’, but need appropriate, accessible and flexible learning opportunities to sustain their CPD. The greatest challenge lies in engaging those most reluctant to embrace the kind of technologies that a coming generation of undergraduates expect as a core part of their day to day learning experience, and there is a significant need for further research to understand the nature (and size) of the gap, and identify best practice.

56


REFERENCES Alexandrou A., Field K., and Mitchell H., (Eds.), (2005), The Continuing Professional Development of Educators: Emerging European Issues, Symposium Books, Oxford. Chivers G., (2005), The training of lifelong learning professionals as researcher, Journal of European Industrial Training, Vol. 30., No. 5, pp. 330-348. Cope S., Bradley S., Crawford K., Rothwell A., (2008), Evidencing CPD: planning for an uncertain future, Conference workshop, Credit-bearing Professional Development in HE Learning & Teaching, Escalate Education Subject Centre, Centre for Excellence in Teaching & Learning (CETL), Ellison Building, Northumbria University City Campus, Newcastle upon Tyne, December 1st 2008 Dunne R., (2005), How Teachers Develop their Teaching, Report commissioned by LTSN Generic Centre, York, http://www.ltsn.ac.uk/application.asp?section=generic&app=resources. Green H, and Hannon C., (2007), Their Space: Education for a digital generation, DEMOS, London Jackson N., (2005), Developing Higher Education Teachers: What works’ Higher Education Academy Seminar September 14th 2005, http://www.health.heacademy.ac.uk/news-events/eventsbox/heayorksept05/, accessed March 6th 2008 King H., (2004), Continuing Professional Development in Higher Education: what do academics do?, Planet, No. 13, pp 26-29 Norris R., (2007), The UK professional standards framework and career-long learning, Academy Exchange, HE Academy, No. 6, Summer 2007, pp 30-31. Pazy A., (1992), The threat of professional obsolescence: how do professionals at different career stages experience it and cope with it?, Human Resource Management, Fall 1990, Vol 29, No.3, pp 251-269 Prensky M., (2001a) Digital Natives, Digital Immigrants Part 1, On the Horizon, September/October 2001, 9, (5), 1-6 Prensky M., (2001b) Digital Natives, Digital Immigrants Part 2, On the Horizon, September/October 2001, 9, (6), 1-6. Rothwell A., and Arnold J., (2005), How HR professionals rate continuing professional development, Human Resource Management Journal, Vol. 15. no. 3, pp. 18-32 Rothwell A., and Herbert I, (2007), Accounting Professionals and CPD: Attitudes and Engagement – some survey evidence, Research in Post Compulsory Education, Vol. 12, No. 1, pp 121-138Rothwell A., Cope S., Bradley S., Crawford K., (2008), What Academics Value; A Comparative Analysis of Research into Continuing Professional Development (CPD) in Four English Universities, Society for Research into Higher Education Annual Conference, Adelphi Hotel, Liverpool, 9-11 December 2008 Salway G., Katz R.N., Caruso J.B., Kvavik R.B., Nelson M.R., (2006), The ECAR Study of Undergraduate Students and Information Technology, 2006, Research Study from the Educause Center for Applied Research, Boulder, Colorado. Sadler-Smith, E. and Badger, B. (1998) The HR practitioner's perspective on continuing professional development, Human Resource Management Journal, Vol. 8, no. 4, 66-75 Sharpe R., Benfield G., Roberts G., Francis R., (2006), The undergraduate experience of blended e-learning: a review of UK literature and practice. Toledo C.A., (2007), Digital culture: Immigrants and tourists responding to the natives’ drumbeat, International Journal of Teaching and Learning in Higher Education, 19, 1, 84-92 Zevenbergen R, (2007), Digital natives come to PreSchool: implications for early childhood practice, Contemporary Issues in Early Childhood, 8, (1), 19-29.

57

ISBN: 978-972-8924-82-9 © 2009 IADIS

INVESTIGATING UNIVERSAL ACCESS FROM A HUMAN DEVELOPMENT PERSPECTIVE Florence Nameere Kivunike, Love Ekenberg, Mats Danielson Department of Computer and Systems Sciences - DSV, Stockholm University, Sweden

F F Tusubira Makerere University, Uganda

ABSTRACT To ensure that the rural poor also benefit from Information and Communication Technology (ICT), governments in developing countries have employed the universal access policy to provide affordable access to communication facilities, within reasonable distances, at an affordable cost. While universal access (UA) is one of the development initiatives aiming to improve the day-to-day lives of people in marginalized communities, little is known about the extent to which UA has catered for and contributed towards human development. This paper addresses this gap by considering human development in terms of Sen’s capability approach which emphasizes the creation of opportunities and freedoms for people to live the lives they value. It specifically reports investigations of Uganda’s Universal access fund - the Rural Communications Development Fund (RCDF) which has registered a lot of success in as far as meeting its infrastructure targets is concerned. The study established that RCD has evolved from infrastructure deployment to service delivery. The apparent next level is the need for a human development approach which provides a beneficiaries’ perspective to universal access deployment, one that explicitly considers the beneficiaries’ needs and values. KEYWORDS Human development, universal access, universal service, rural communications development, Uganda

1. INTRODUCTION Universal Access, broadly defined in this paper as the provision of convenient and affordable access to communication facilities by people in marginalized communities, is a key policy objective in developing countries. While several universal access efforts have registered success in as far as infrastructure deployment is concerned, there is still limited evidence on how the infrastructure has benefited the rural poor in terms of human development. This indeed comes as no surprise, because the initial conception of universal access was based on the simplistic assumption that the provision of the means of access at affordable cost would lead directly to economic benefit and human development. The policy formulation and strategy implementation consequently tended to treat the target beneficiaries as passive spectators rather than active players, running contrary to accepted development theories. This creates the environment for apparent implementation success that however fails to achieve the underlying policy objective. This kind of failure is an example of the role development theories play in the effective delivery of development initiatives. Prakash & De’(2007) also established that a land reform project in India that was aimed at improving access to cultivation land by farmers within the community ended by actually marginalizing the smaller farmers because they could not participate in the markets. This was because the policy adopted neo-liberal tendencies, increasing efficiency in the land markets by computerizing and centralizing the issuance of land records, and therefore effectively created investment opportunities for transnational companies Particularly the current quest for approaches to enable the rural poor benefit from Information and Communication Technologies (ICT) brings to bear the notion of human development. Human development, proposed by Sen’s Capability approach as a process of expanding people’s capabilities or choices1, provides 1

Human development report 1990 - http://hdr.undp.org/en/reports/global/hdr1990/

58


for the explicit consideration and focus on the intended beneficiaries. Furthermore it is increasingly argued that failure of ICT4D initiatives in terms of none or limited use is attributed to the limited conception or understanding of the development the initiative aims to achieve or improve (Prakash & De’, 2007). This gives the human development concept a central role in the meaningful design and implementation of universal access funds in order to benefit people in rural communities in developing countries. This paper investigates the extent to which Universal Access Funds have taken cognisance of the human development concept. This is achieved through a general review of literature and discussion of the human development concept and universal access funds in sections two and three respectively. Sections four through six present the case of UAF in Uganda, from policy to infrastructure implementation and actual use as well as achieved benefits by the intended recipients. Then section seven discusses the UAF in Uganda in relation to human development and in conclusion section eight makes suggestions for human development-centered universal access implementation. It is assumed that this will facilitate an understanding of the varied impacts and assist governments in planning or revisiting the universal access/service policies to benefit their intended beneficiaries.

2. HUMAN DEVELOPMENT CONCEPT – CAPABILITY APPROACH Human development presents a paradigm shift in development thinking whose prominence arose from the introduction of the United Nations human development report in 1990. Unlike approaches which emphasize development as the lack of income or economic growth, human development adopts a multi-dimensional, pluralistic perspective which also considers the value of other aspects like health and education to one’s life (Kuonqui, 2006; Sen, 2000a). Furthermore, human development is people-centered and reiterates the importance of human beings as both a means and an end, unlike earlier approaches that regarded human beings as means of production and economic growth (Sen, 1990). The concept was greatly enthused by Amartya Sen’s works in welfare economics, social choices, poverty but most specifically his capability approach. This is particularly evident from the definition it adopts - expansion of capabilities or choices that people value. In addition researchers and practitioners are actively involved in the human development and capability association applying the capability approach to human development research and practice2. In affirmation Kuonqui (2006) and Muro & Tridico (2008) point out that while human development is an advocacy instrument, the capability approach provides its analytical basis; an aspect that this section adopts in proposing an evaluative space to investigate the influence of universal access on people’s lives. The three core concepts central to the capability approach are functionings, capability and agency. Functionings are the “things one values doing or being” (Sen, 2000b pp 75), which comprise an individual or group’s well-being once achieved. Capability or capability set on the other hand is a combination of potential functionings - opportunities, also referred to as the freedom one has to achieve a life they value or have reason to value (Sen, 2000b pp 87). The approach further highlights the important influence of contextual factors on the conversion of commodities into functionings and opportunities. For example in addition to providing infrastructure it is worth establishing whether the intended beneficiaries have the relevant skill to exploit the infrastructure, and how this affects the opportunities they can derive from the infrastructure. Sen further points out that besides providing opportunities, agency – a person’s ability to pursue and realize goals in relation to their personal needs and aspirations is essential for development. From this perspective beneficiaries are actively involved (participate) in the development process explicitly addressing empowerment and sustainability which are central concerns to human development. It is evident here that the capability approach shares the general concern that failure to design, implement and evaluate development programmes from the beneficiaries’ point of view can result into failure (Schischka et al., 2008). According to Sen enhancing people’s freedom through the provision of capabilities/ opportunities to live the lives they value is the main objective, both as a means and an end to development (Sen, 2000b). Given the multidimensional characteristic, Sen particularly proposes five interrelated instrumental freedoms valuable for development including: economic facilities, political freedoms, social opportunities, transparency guarantees and protective security (Sen, 2000b).

2

http://www.capabilityapproach.com/

59

ISBN: 978-972-8924-82-9 © 2009 IADIS

3. UNIVERSAL SERVICE AND ACCESS Both Universal Service (US) and Universal Access (UA) are policy objectives which originated from the need for universality in the provision of telecommunication services. While universal service emphasizes ‘everyone’ in terms of the availability of connectivity to public telecommunication networks by individual households nation-wide, universal access focuses on ‘everywhere’ by referring to shared, communal access to publicly provided ICT facilities and services (Dymond & Oestmann, 2004; Intven & Tetrault, 2000). Central to both concepts is the availability, affordability and sometimes the quality of services of telecommunication infrastructure and services (Verhoest, 2000). Fundamentally the ultimate goal of UA strategies is US, achievable with universal access either as a precursor to or a subcomponent of universal service. As a precursor, UA in terms of the availability of easily accessible telecommunication and ICT services at affordable costs on a communal or shared-basis implies that everyone has been catered for even at household level. On the other hand UA as a subcomponent of US means that US can in certain instances precede the achievement of universal access; an aspect that has also been referred to as leap-frogging. This is most evident in the growth and penetration of mobile telephony in the developing countries which has enabled individuals (and even households) to acquire mobile phones in places where publicly accessible services are not possible or not yet in place. Similar to affordability and availability, UA and US is also conceptualized in terms of the “access gaps” basing on two overlapping dimensions: poverty and isolation (Navas-Sabater et al., 2002). This model has especially been influential in defining, implementing and determining the funding approaches to universal access. In this respect two access gaps exist; the market efficiency gap (MEG) and true access gap (TAG). MEG refers to the difference between the achievable level of service penetration given restrictive market conditions, and the level of service penetration expected under optimal market conditions (Navas-Sabater et al., 2002). Addressing this gap aims to provide services for especially the urban poor who cannot afford private facilities which involves setting favorable market-oriented policies for the private sector (operators and entrepreneurs) to provide and profit from telecommunication services. On the other hand the true access gap is a result of both poverty and isolation, typical of a rural community. This context is considered not commercially viable which requires special interventions in form of incentives such as subsidies to encourage service providers from two approaches: either one-off grants for deployment of infrastructure through Publicprivate partnerships aimed to reduce capital investment, or vouchers to proprietors to enable them meet the cost during a transitional period. These subsidies are sourced from development partners and levies from national and private telecommunications operators. It is clear from the discussion so far that the current UA strategies and approaches have limited consideration of human development in as far as multidimensional and people-centered aspects are concerned. First, poverty in this context is one-dimensional perceived only in terms of an individual’s or community’s disposable income, unlike the pluralistic CA concept which considers all instrumental freedoms. This has resulted into the misconception that the rural poor cannot afford ICT services, one of the major guiding principles in designing UA. Secondly, while UA targets the rural poor, it is evident from the discussion above that focus is more on creating efficient markets and conditions for the service providers rather than meaningful services for the target beneficiaries. Additionally, the target beneficiaries seldom participate in the UA policy and strategic planning process. As a result UA is mostly limited to infrastructure provision and little on content which would preferably depend on people’s information needs. It is important to underscore that while the rural poor might not know the social, economic and political opportunities that they could obtain from ICT, they know what they need and value in life. A human development approach to universal access would therefore attempt to address this gap which would subsequently ensure the ownership and sustainability of such facilities. The general assumption put across and one that will be further tested through the investigation of UA in Uganda in the subsequent sections is that approaching universal access from strategies that do not explicitly consider a people-centered, human development perspective will result into failure and limited benefits of the intervention to rural development.

60


4. METHODOLOGY The aim of this paper is to establish the extent to which human development in terms of life’s opportunities that people value - an inherent goal to development initiatives has been catered for in universal access implementation and use. To meet this objective, an empirical study was conducted to investigate UA policy and implementation in Uganda. This study involved two aspects: a) a desktop study of strategies, policy documents, review reports as well as online resources on UA implementation in Uganda; and b) field visits to two rural internet access points that had benefited from the universal access fund. Informal, semi-structured interviews were conducted with randomly selected users who were using the facilities at the time of visiting. The interviews sought to ascertain the general perception of the services from the users’ standpoint and the extent and purpose of usage. Interviews were also conducted with the facility managers to establish the purpose and process through which the services were implemented. It is worth pointing out that while universal access involves the provision of a number of ICT services, the field visit specifically focused on the internet facilities on the assumption that anyone who found value in using internet had good knowledge of other basic ICTs like payphones. From a human development perspective, the predominantly qualitative study particularly sought to establish the extent to which universal access meets the following capability approach principles: • Agency, participation – the extent to which target beneficiaries are involved in their development; how community/people’s needs are taken into consideration; • Multidimensionality - whether the implementation of UA efforts consider the wide range of people’s freedoms; • Contextual factors – the extent to which the surveyed UA efforts consider the influence of the social conditions such as level of education, income etc on the adoption and use of the provided services.

5. UNIVERSAL ACCESS POLICIES AND STRATEGIES IN UGANDA Given that UA specifically targets the rural communities, in Uganda universal access is referred to as Rural Communications Development (RCD). The RCD policy (RCDP) was developed by the national telecommunications regulatory body, the Uganda Communications Commission (UCC) in 2001. According to the RCDP whose implementation did not start until 2003, RCD initially aimed “to ensure that basic communication services of acceptable quality [were] accessible, at affordable prices, and at reasonable distances, by all people in Uganda (UCC, 2001 pp. 7)”. To address the availability and accessibility challenge, the policy prioritized the provision of public telephony and postal services in each sub-county as well as Internet points of presence with at least one public Internet café in all districts. The policy further explicitly considered aspects like awareness and literacy which would affect utilization by supporting the setup of training centers and developing web portals for each district. To ensure affordability, RCDP and UCC advocated for the promotion of competition among service providers as a means of reducing tariffs. Furthermore the general principles upon which the RCD Fund (RCDF) was established include: assisting areas where commercial services were not feasible, providing basic communication services and leveraging investment rather than provide all solutions. In addition funding is accessed through competitive bidding and the fund is contributed to by service operators (UCC, 2001 pp. 7-8). Besides the 1% levy on operators’ gross annual revenues, financial support has been obtained from development agencies such as the World Bank and the International Telecommunications Union (ITU). With an overall aim of long-term sustainability of services, funds are competitively awarded through ‘smart subsidy’ – a one-time start-up incentive to the interested bidders (UCC, 2005a). Bidders are further required to demonstrate the potential to sustain RCD initiatives after the start-up subsidy period expires. After a series of telecommunication sector performance reviews, recommendations for revision in the overall sector policy were made in 2005 to explicitly adopt a human development perspective (UCC, 2005b). This was borne from the realization that after the initial regulatory period there was a low level of ICT penetration and accessibility was still evident nation-wide, as well as the need for a holistic approach towards national development. From a human development perspective, the new policy recommendations generally expressed the need to target sector-specific ICT goals. Universal access was defined by the following question: “What infrastructure and services must be in place (where and when) in order to enable the human

61

ISBN: 978-972-8924-82-9 © 2009 IADIS

development plans and objectives of the different sectors?”(UCC, 2005b pp 87) The new objectives were defined in terms of infrastructure to provide accessibility and availability throughout the country to a minimum of a voice and data network point by the year 2010 in terms of institutional and public data and voice access points and ensuring affordability by the marginalized populations through various pricing approaches (UCC, 2005b pp 90 - 91). The policy further stresses the need for relevant, timely and accurate content to meet people’s information needs and alludes to the need for creating awareness and the introduction of ICT skills to enable people exploit ICT beyond social benefits. Although these recommendations have not been formally adopted, they are currently guiding RCD implementations. To date RCDF has registered successful infrastructure deployment actually meeting its targets way before the set deadlines (RCDF, 2007; UCC, 2005b). It is also being used as a best practice for universal access deployment in other developing regions (e.g. Dymond & Oestmann, 2004). To provide sufficient ground for a critique on the extent to which the RCD policies and strategies incorporated human development concerns, the next section discusses the perceptions of RCDF from the beneficiaries’ perspective.

6. FIELD STUDY FINDINGS - RCDF FROM THE BENEFICIARIES’ PERSPECTIVE With the overall aim of establishing how people in the target communities perceived the RCD services, field visits were conducted at two sites: Nangabo vocational training institute and Busoga Rural Open Source and Development Initiative (BROSDI) in February 2008. In total 13 participants were interviewed at both sites; at Nangabo vocational institute the principal, 2 teachers and 2 students were interviewed, while the manager and 7 users were interviewed at BROSDI.

6.1 Nangabo Vocational Training Institute The vocational training institute is located in Nangabo sub county 14km north of the capital city. The community is classified as rural consisting of about nine villages and 12,000 households. While some residents commute to the city for work, the major activity is subsistence farming. In addition to the vocational training institute the sub county has a number of both primary and secondary schools. As a result the community is mostly semi-illiterate, with an elementary ability to read and write. The institute has a total of 60 students undertaking courses in nursery school teaching, catering, tailoring, fabric weaving and design and basic computer training for those who can afford an extra fee. The main purpose for incorporating computer facilities at the institute was to provide computer literacy to the students and community. The facilities which were initially set up in 2000 currently comprise 23 standalone computers, a photocopier and printer. Only one of the computers has Internet access. The computers have mostly been donations from friends, RCDF and Uganda Telecom Ltd (one of the telecommunications service providers). It is worth noting that at the time of this field visit students were just reporting back to school, as such only 2 students were available for interviewing. The Principal reported that facilities were mostly used by students at the institute, those within the community and schools. The use of facilities was limited to basic secretarial services like: a) basic training in typing, sending email, and search/find information on the Internet; b) typing, printing and photocopying exams for schools within the community; c) writing and printing letters; d) some (literate) members of the community use the internet and e) conducting courses and workshops for tutors to obtain basic training in computing. For purposes of financial sustainability, users from the community were charged UGX 1000 (USD 0.52), while students enrolled at the institute paid UGX 40,000 (USD 21) to get trained in computer skills per school term (three months). The integrated and creative use of ICT to assist in instruction delivery or enhance the quality of the study programmes was rather limited. For instance the Principal pointed out that computers were only used to do assignments in computing, as such computer facilities were an add-on rather than a support to the existing services. While the majority of users (both tutors and students) were not aware of this possibility, use by those who were aware was affected by the slow internet connection and power outages. The facilities were actually not in use at the time of visiting due to a power failure in addition to the institute not having any

62


power backups. In addition, some of the students did not use email because they had no one to communicate to, while others could not afford the extra USD 21 to benefit from the training. Obtaining technical support and maintenance of the equipment was also a great challenge, although the institute was obtaining free support and maintenance from a firm in town.

6.2 Busoga Rural Open Source and Development Initiative (BROSDI) BROSDI3 headquarters are located in Baitambogwe sub county Mayuge district in Eastern Uganda, on the shores of Lake Victoria, about 145km from the capital city. The community setting is typically rural, characterized by fishing and subsistence farming as the major economic activities. Mayuge is a new district that was curved of Iganga District in 2000. It has since had several initiatives geared towards social and economic development such as developing the road network, training farmers in income generating activities and sensitization on responsible fishing practices. The district has about 137 government aided primary schools, a few private secondary schools and hardly any tertiary institutions. According to the district website the literacy rate was at 45% in 2005 and schools were faced with high drop out rates4. BROSDI is one of the development initiatives that was setup to improve the lives of people in the surrounding communities through the provision and sharing of information relevant to their livelihoods. The center comprises a computer lab, resource library and farming demonstration sites. The aim of setting up the computer facilities was to enhance the traditional communication and information sharing practices through the use of ICT. The computer lab has 12 networked computers all connected to the Internet. The manager reported that access to Internet was free for the first 30 minutes after which a user was required to pay for extra time and as a result approximately 30 people used the computer facilities daily. However this increase in use was registered only after users were granted free access to the services. The facilities were mostly used in the afternoons after farmers had come back from the fields. In addition to the day-to-day running of the facilities, the manager was also responsible for basic maintenance and conducting the training sessions. On realizing the potential benefit of the facility to people’s lives, BROSDI devised various means to ensure a constant cash flow like sourcing funding through other projects, and requiring university students and working middle class pay for services. Furthermore to cater for both the literate and illiterate within the community, BROSDI was working with the Ubuntu5 to acquire a browser in one of the local languages. Interviewed users pointed out that they used the facilities for: training; searching for information on the Internet; blogging, text messaging using the Web 2.0 tools; watching movies/clips; reading local, international and sports news; communicating with family and friends through email. The center was also has registered members and works with extension farmers who collect and disseminate information on farming/agricultural practices, health etc through the BROSDI websites. Furthermore, a secondary school student pointed out that he conducted some research though this was minimal. Apart from the ICT related activities, the center also conducts other activities like: youth workshops on health related matters; education programs for children in upper primary classes over the weekends; farmers also occasionally meet at the center to share information in relation to good farming practices e.g. a local technique in fighting crop pests. In terms of the general impact of the center on people’s day-to-day lives, participants pointed out that by using the services they were able to save on transport costs and time, some students acquired scholarships through the information they obtained from the Internet, farmers obtained valuable information on improving their produce as well as a general improvement in primary students’ performance within the community. Despite the registered benefits there were still some challenges that constrained use: electricity power outages and fuel expenses in using generator power backup, users had to walk long distances to the community center since it was the only one within the area then and the slow connectivity especially frustrates the students.

3 4 5

http://www.brosdi.or.ug, http://www.celac.or.ug, http://360.yahoo.com/hope_childrens_club http://www.mayuge.go.ug http://www.ubuntu.com/

63

ISBN: 978-972-8924-82-9 © 2009 IADIS

7. DISCUSSION This section seeks to analyse and discuss the extent to which the RCD policies, strategies as well as implementation in Uganda are designed from a human development perspective. It further establishes how the RCD efforts have benefited the rural population from a human development perspective. This is based on the premise that approaching rural communications development from a human development perspective, both as a multi-dimensional and people-centered concept is the optimal option to ensure meaningful deployment of ICT in the rural communities.

7.1 RCDP and Strategies in Relation to Human Development From a strategic standpoint, Uganda appreciates the connection between human development and rural communications development. However, while the approach at policy level was improved – i.e. relating rollout requirement to the macro human development plans, it is still deficient in that it did not take an explicit human development approach. The overall driver was service delivery, rather than response to the micro-level development needs of communities. In actual fact the significant difference between the old and new policy objectives is pinning infrastructure deployment to specific sectors; however affordability, relevant content and creating awareness remained essential in both deployment phases. In relation to CA’s conception of human development, the new RCD policy is limited in as far as the active role of the target beneficiaries is concerned. This may be based on the assumption that people in rural communities are not aware of the ICT potential towards their day-to-day lives. While this may be true, people in these communities know their needs which should be the basis of ensuring the provision of relevant ICT services. For example the district web portals whose target was to meet the information needs of the communities remain under- utilized because the information provided does not address the needs of the people in the different communities. However this information is more relevant to potential investors interested in obtaining demographic characteristics of the different districts. This is typical of a top-down approach to ICT service provision which affects ownership and the eventual sustainability of ICT services in the target communities. Furthermore, it is evident that the deployment of universal access (RCD) involves several stakeholder groups with differing implementation goals which sometimes affects the nature and quality of service. While the government (i.e. regulator) as the visionary whose goals as expressed by policy may to some extent target the welfare of the rural population, the implementers (proprietors, service providers, operators) have varied goals and expectations determining the nature of services they deliver to the target beneficiaries. For instance while RCD considered the need for ICT literacy and awareness at policy level, most of the facilities set up as training centres to address this need quickly converted to commercial internet cafes. Although policy requires proprietors to exhibit financial sustainability of services, a similar caveat on the nature and quality of services delivered would assist in ensuring that the set goals are followed through to service delivery. Finally, studies and anecdotal evidence show that the rural population is willing to pay for ICT services even in places where they appear not to be commercially viable once people appreciate their inherent value (Kivunike et al., 2009; Shirley et al., 2002). Evidently while affordability is a key determinant of accessibility and utilization of ICT services, there is need to emphasizing the relevance, exposing the value of the ICT to people’s needs. A good example is the mobile phone whose penetration and use are on the increase due to its ease of use and clear benefits, despite the high tariffs. Therefore while market favorable strategies to ensure affordable services like competition, privatization etc have been proven to expand accessibility and improve service quality; it is imperative that policy explicitly caters for the value of ICT services towards the target beneficiaries. This will encourage acceptance and use which in turn addresses the affordability-sustainability challenges. It is important to underscore that in the Ugandan context the implementation of the RCD programme from a human development perspective is still in its infancy to evaluate performance. However assessing performance of the initial RCD phase is vital as baseline and means of policy refinement of the new RCD programme as discussed in the following section.

64


7.2 RCD Implementations in Relation to Human Development It is clear from the Nangabo vocational training institute case approaching RCD from a non-human development perspective is a potential source of failure. First, although the aim in setting up the computer facilities was to serve both the Institute and the Community, the target beneficiaries were not consulted on the nature and purposes of services. As a result the acquired facilities did not explicitly meet the needs of either beneficiary. Furthermore, the lack of skilled personnel and knowledge of how ICT could integrated to support the institute’s programmes meant that the facilities were alienated from the study programmes and could not be effectively used in this respect. Subsequently this under-utilization has affected the sustainability, ownership and expansion of facilities. Furthermore, the requirement for students enrolled at the institute to pay an extra fee so as to acquire basic computer skills is source of inequalities among those who can afford and those who cannot afford. On the affirmative though the acquired skills expand people’s capabilities giving them the possibility to obtain ICT-related jobs. In contrast, the approach adopted by BROSDI is characteristic of a human development approach. The proprietors in this case adopted an approach that considered contextual influences on ICT use in addition to ensuring a multi-dimensional influence on people’s lives. This is apparent in the partnership with Ubuntu to develop a browser in the local languages catering for those who did not know English and providing activities for both the young and old in the community etc. It is further evident in this context that the proprietors realised that ICT in isolation cannot benefit its target beneficiaries and it was therefore introduced to facilitate already existing information sharing activities. The centre further adopted a people-centered approach to service provision which was based on the needs of the community. For instance weekend classes for primary school going children were introduced after realizing that the district had poor performances as well as high drop out rates. Furthermore the ability to produce and share content through websites and discussion forums has empowered farmers with various skills. These aspects coupled with free Internet access and various funds generation activities account for both the social and financial sustainability as well as local ownership of the facilities. This further confirms the premise that indeed a human development perspective is essential for the effective deployment of rural communications development. A down-side to BROSDI’s approach is cited in offering free services which is not sustainable on the long term. Furthermore, while the centre found it appropriate to translate the browser in the local languages, majority of the content is still in English which will also limit use by those who do not know English. On the other hand though both cases differed in the approach to service delivery, full exploitation of facilities was limited by the slow internet connectivity. Furthermore it can be deduced from the findings in both cases that the rural poor still require an incentive to exploit ICT for their well-being. For instance, although BROSDI tailored their services to the needs of the community, they still had to provide a free slot to encourage people to appreciate and use the service. As one of the interviewees rightly pointed out; “using ICT to leverage development requires innovative approaches which will make the technology relevant for the needs of the rural poor”.

8. CONCLUSION It is vital to point out that the analysis and conclusions drawn in this paper are based on a literature survey of policy documents and only two cases. Therefore the findings might not be generalizable to RCD nation-wide, but provide valuable insight for further deployments. It is manifest from the above discussion that the universal access efforts in Uganda have evolved from basic infrastructure to service delivery. In summary these efforts are characterised with limited participation by intended beneficiaries in service development and delivery, a one-dimensional definition of poverty and hardly any consideration of the social factors that influence service adoption and use. The apparent next level is adopting a beneficiaries’ perspective to RCD deployment by explicitly focusing on aspects of participation, multidimensionality and contextual influences on ICT adoption which will ensure that people’s needs and values are central. In this respect RCD efforts should always focus on answering the question, development for whom? Similarly and in addition to ensuring affordability, effort should go into empowering citizens to establish the potential ICT benefits to their day-today lives. Finally, there is a need to bridge the gap between policy and implementation, because while policy considers aspects of human development, implementation adopts varied approaches which affect the nature

65

ISBN: 978-972-8924-82-9 © 2009 IADIS

of services delivered. This can all be summed up by a quote from Olukoshi (2007) cited in (Prakash & De’, 2007pp. 276) “The desire for technical solutions to development problems should not take on a life of its own where we forget that development is about people and what they think and how they feel matters”.

REFERENCES De', R. 2006. Evaluation of E-Government Systems: Project Assessment vs. Development Assessment. EGOV 2006. pp 317-328. Dymond, A., & Oestmann, S. 2004. Rural ICT Toolkit for Africa. Washington, DC: African Connection and infoDev/World Bank. Intven, H., & Tetrault, M. 2000. Module 6 - Universal Service. In infoDev (Ed.), Telecommunications Regulatory Handbook + module 6. Washington DC, USA: The World Bank. Kivunike, F. N., Ekenberg, L., Danielson, M., & Tusubira, F. F. 2009. Examining Contextual Factors that Influence ICT Adoption in Rural Communities in Uganda. In I. I. I. M. Corporation (Ed.), Accepted for IST-Africa 2009 Conference. Kampala Uganda. Kuonqui, C. 2006. Is Human Development a New Paradigm for Development? Capabilities Approach, Neoliberalism and Paradigm Shifts. Paper presented at the International Conference “Freedom and Justice” of the Human Development and Capability Association (HDCA). , Groningen, Netherlands. Muro, P. D., & Tridico, P. 2008. The Role of Institutions for Human Development. Unpublished Working Paper. University of Roma Tre. Navas-Sabater, J., Dymond, A., & Juntunen, N. 2002. Telecommunications and Information Services for the Poor: Toward a Strategy for Universal Access. Unpublished Discussion Paper. The World Bank, Washington, DC. Olukoshi, A. 2007. From Colonialism to the New Millennium and Beyond. IDS Bulletin, No. 38, pp 20-25. Prakash, A., & De’, R. 2007. Importance of development context in ICT4D projects: A study of computerization of land records in India. Information Technology & People, Vol. 20, No. 3, pp 262-281. RCDF 2007. MPs study visit to the Kabarole MCT. Uganda Communications Commission. Schischka, J., Dalziel, P., & Saunders, C. 2008. Applying Sen's Capability Approach to Poverty Alleviation Programs: Two Case Studies. Journal of Human Development, Vol. 9, No. 2, pp. 229-246. Sen, A. 2000a. A Decade of Human Development. Journal of Human Development, Vol. 1, No. 1, pp. 17-23. Sen, A. 2000b. Development as Freedom (Thirteenth Ed.) New Delhi: Oxford University Press. Sen, A. (Ed.). 1990. Development as Capability Expansion. London: MacMillan. Shirley, M. M., Tusubira, F. F., Gebreab, F., & Haggarty, L. 2002. Telecommunications Reform in Uganda. Unpublished Policy Research Working Paper, 2864. The World Bank, Development Research Group. UCC 2001. Rural Communications Development Policy for Uganda, July 2001. . Uganda Communications Commission. UCC 2005a. Funding and Implementing Universal Access, Innovation and Experience from Uganda. Kampala: Fountain Publishers/IDRC. UCC 2005b. Recommendations on Proposed Review of the Telecommunications Sector Policy. Kampala: Uganda Communications Commission. Retrieved 2nd June 2008 from http://www.ucc.co.ug/UgTelecomsSectorPolicyReview_31_Jan_2005.pdf Verhoest, P. 2000. The myth of universal service: hermeneutic considerations and political recommendations. Media Culture Society, Vol. 22, No. 5, pp 595-610.

66


THE HARE AND THE TORTOISE: A COMPARISON OF NORTHERN AND SOUTHERN HEMISPHERE FINDINGS OF THE SITES 2006 Seugnet Blignaut and Christo Els School for Continuing Teacher Education North-West University, Potchefstroom Campus

ABSTRACT This paper reports on a secondary data analysis of the International Association for the Evaluation of Educational Achievement’s (IEA) Second International Information Technology in Education study (SITES 2006) cross-sectional comparison of how Information and Communication Technologies (ICTs) change teaching and learning practices in secondary schools worldwide in the 21st century. South Africa and Chile were the only southern hemisphere countries that participated in the study, along with twenty other countries from the Northern hemisphere. In spite of a successful data collection, ensuring that South Africa, for the first time in a SITES study, achieved the statistical criteria and appeared in the main tables of the international report, the percentage frequencies indicate sizeable differences between the two southern hemisphere countries, as well as between the northern hemisphere countries in terms of achievement of ICT integration into schools. Although this secondary data analysis, using Cramer’s effect size, indicates that many differences are not of practical significance, South Africa lags behind in many aspects. This paper focuses on the discrepancies found in integration of ICTs in classrooms, provision of ICT support, and teacher professional training. KEYWORDS SITES 2006; developing context; survey study; computer literacy; comparative study; teacher professional development.

1. INTRODUCTION South Africa participated in all three Second Information Technology in Education Study (SITES) modules between 1999 and 2006 under the auspices of The International Association for the Evaluation of Educational Achievement (IEA) (Pelgrum et al, 2008:8). The study consisted of three separate modules and explored the relationship between Information and Communication Technology (ICT) and educational change. The first module of SITES, a large quantitative study carried out between 1997 and 1999, was an international cross-section survey of principals and technology coordinators in schools of 26 countries. Module 1 focused on ICT resources and the extent to which schools have adopted and implemented pedagogical practices essential to education in the information society (Pelgrum et al, 2008:5). The second module of SITES was an international qualitative study of innovative pedagogical ICT practices, conducted between 2000 and 2002 in 28 countries according to a selection of mutual criteria that identified 174 innovative classrooms. National research teams followed case study methods to collect data on the pedagogical practices of teachers and learners, the role that ICT played in these practices, and the contextual factors that supported and influenced them (Pelgrum et al, 2008:7-9). The third module of SITES in 2006 collected data from 22 countries. Three questionnaires targeted mathematics and science teachers, principals and technology coordinators to determine teachers’ pedagogical practices and use of ICT in teaching and learning. This survey probed respondents on their perceptions of important school or system level factors and the current role of ICT to support teaching and learning. Twenty countries from the northern hemisphere and two countries from the southern hemisphere, South Africa and Chile, participated in SITES 2006. The international report (Law et al, 2008b) indicates that in 2006, South Africa lags behind northern hemisphere countries in most respects. In spite of South Africa’s relative strong African economy, and its remarkable spending of about 28% on education annually, Chile, also a developing country, outperforms South Africa. This paper aims to highlight some of the aspects where

67

ISBN: 978-972-8924-82-9 © 2009 IADIS

South Africa should put improvement plans into action, as well as to explain the background to some of these discrepancies. When examining the extent of the differences between the North and the South, Aesop’s fable of the unequal race between the hare and the tortoise springs to mind.

2. BACKGROUND South Africa comprises 47.4 million people of diverse origins, cultures, languages and beliefs. About 79% are black (African), 9% white, 9% of mixed race and 2.5% of Indian descent. The surface area of 1.2 million square kilometres is divided into nine provinces and about half of the population lives in rural and semi-rural areas. South Africa is a multilingual country that recognises eleven languages as official. Most of the languages are indigenous to South Africa and around 40% of the population speaks either isiZulu or isiXhosa. IsiZulu is the home language of 23.8% of South Africa’s population, followed by isiXhosa at 17.6%, Afrikaans at 13.3%, Sepedi at 9.4%, with English and Setswana at 8.2% each (South Africa.info 2006:online). SITES2006 determined that language is an obstacle for South African schools for ICTimplementation in teaching and learning, as English is not the primary home language. English is the most common language used on the World Wide Web and it is therefore not surprising that the majority of the respondents saw language as a problem (Anderson et al, 2008:53). South Africa faces considerable education challenges on whole and in particular with integration of ICT into schools. Class sizes in South Africa are steadily increasing, but resourcing of classrooms remains a challenge. The Department of Education’s National Education Infrastructure Management System quantifies the physical infrastructure for education in all schools in South Africa. The NEIMS 2007 assessment report indicated that of the 25,145 public ordinary schools, 57% comprise 30-34 students per teacher, and 5% have more than 45 students per teacher (Department of Education 2007:online). The concept of “students per teacher” is highly contested in South Africa, and the concept of “class size” is a more accurate gauge. During 2003, the TIMSS study encountered average class sizes of 45, and often up to 75 grade eight students (Gonzales et al, 2004:online). A consequence of large classes is a decrease in available teaching and learning facilities and resources as well as overcrowded classrooms. Many teach in dilapidated classrooms with insufficient furniture, space and equipment (Onwu 1998:120). The NEIMS 2007 report indicates that in 56% of the 1,972 listed schools, more than 10% of students are without desks, and in 58% of schools, more than 10% of students are without chairs. Under such conditions, teachers cannot realize 21st century pedagogical goals unless information technology equipment and tools are available to them. South Africa was the only country in the SITES 2006 study that could not provide students with full access to computers. The percentage of schools with computers increased from 18% during the 1998 study to about 38% during the 2006 study, despite South Africa’s enormous development leap during this period (Pelgrum 2008:74). Table 1 lists South Africa’s need for ICT equipment and applications. Table 1. Percentage frequencies of schools in which ICT equipment are needed but not available * Needed ICT equipment and applications** South Africa Chile Country with lowest % ICT equipment 83.12% 51% Singapore 2% Tutorial software 88.49% 54% Slovenia 5% General software ( word processing, spreadsheet and presentation 64.30% 9% Canada, Norway, Hong software Kong, Singapore 0% Multimedia production tools 90.87% 43% Hong Kong 3% Simulation 93.05% 49% Singapore 24% Communication software 83.00% 19% Singapore 0% Digital resources 79.64% 26% Denmark 6% Digital e-mail accounts 84.04% 39% Slovenia 5% * (Pelgrum 2008:83) ** SITES 2006 Technical questionnaire frequencies for items 4A, 4B, 4C, 4D, 4F, 4G, 4H and 4M, respectively

68


3. METHODOLOGY The National Research Coordinators (NRCs) obtained permission from the national and provincial Departments of Education for administering the three survey instruments to sampled schools in the nine South African provinces. During the national pilot, the NRCs and co-researchers encountered minimal problems with the instruments in terms of local language or comprehensibility. They determined that submitting the survey instruments according to a personal interview was the only viable strategy for the South African context. Due to insufficient internet access, online data collection (ODC) is mostly unavailable. To administer a single survey instrument took about an hour, but it precluded missing data and fostered a high completed-survey return rate. The NRCs trained Masters-level students as fieldworkers with thorough theoretical grounding for quantitative research, as well as for challenging situations in the field through role playing sequences. A Geographical Information System (GIS) matched the sampled schools to the locations of the 102 fieldworkers who participated in the South African data collection. Administrative personnel contacted each school to determine the school’s location, ascertained that the school contained Grade 8 mathematics and science teachers, made appointments with the principals, identified teachers for the survey interviews, and reconfirmed fieldworker appointments at the schools. The NRCs remunerated fieldworkers who presented fully completed survey instruments for their effort and travelling costs. The IEA Data Processing Centre sampled the South African participating schools from the Higher Education Management Information System (HEMIS) database. A stratifying technique divided schools with Grade 8 mathematics and science learners into two sub-strata consisting of computer-using schools and noncomputer-using schools. From this, 504 randomly selected schools and four replacements schools comprised the sample of the main South African study (Law et al, 2008a:27-29). Four replacements schools substituted schools that did not exist, where fieldworkers could not access the assigned schools, or schools that did not honour appointments with the fieldworkers. The South African data collection took place during October to November 2006. The chosen mode of data collection paid off. On average South Africa obtained an instrument return rate of more than 90% — far more than the IEA required return rate of 85%. The IEA commended the South African team, recommended this approach for use in other developing contexts, and acknowledged the quality of the fieldworker training and data integrity. The report of the international study became available during March 2008 (Law et al, 2008b). This paper considers a secondary analysis of the international data, focusing on the position of South Africa compared with Chile, the other southern hemisphere country, as well as with the northern Hemisphere countries. When comparing the percentage frequencies of South Africa and Chile, it seems that Chile outperforms South Africa in many respects. The abundance of available data motivated the authors to perform a two-way statistical analysis to calculate practical significant differences between South Africa and Chile with the Cramer’s V as effect size. An effect size V ≥ 0.5 is considered a large effect with practical significant difference; an effect size of 0.3 ≤ V ≤ 0.4 a medium effect that tends towards practical significance difference; and an effect size of V ≤ 0.2 a low effect with no or very little significance.

4. RESULTS Although many variables influence the use of ICTs in teaching and learning, many list classroom integration, technical support available to teachers, and teacher training as some of the most important factors for the successful and sustainable integration of ICTs into the curriculum.

4.1 Integration of ICTs into Classrooms The availability of computers is one of the access related considerations of a school’s infrastructure. In the 1980s, the major problem of integrating ICT effectively into education was insufficient hardware and inappropriate software available to teachers. SITES 2006 indicates that while approximately 38% of South Africa’s schools had computers, 96% of Chile’s and 100% of Finland’s schools had computers available for teaching and learning. In 2006, only 18% of computers in South African schools connected to the Internet. Table 2 illustrates the location of the computers throughout schools. Other ICT-equipment (e.g. laboratory

69

ISBN: 978-972-8924-82-9 © 2009 IADIS

equipment, musical instruments, overhead projectors, slide projectors and calculators) was available to 75% of SITES 2006 schools. An exception was South Africa at 17% (Pelgrum 2008:80). These figures reflect on the generally low and unequal resourcing of South African schools, especially in rural areas. Table 2. Percentage frequencies indicating where computers were located in schools according to the technology coordinators * Location in schools ** South Africa Chile Country with highest Most classrooms 0.80% 1.0% Hong Kong 69% Some classrooms 3.5% 3.1% France 70% Computer laboratories 49.3% 98.0% Singapore 100% Library 9.8% 25.2% Hong Kong 95% Other places 10.1% 10.9% Hong Kong 66% * (Pelgrum 2008:89) ** SITES 2006 Technical Questionnaire frequencies for items 9A, 9B, 9C, 9D and 9E, respectively

Table 3 indicates an array of activities that mathematics and science teachers use ICTs in their teaching and learning. In general, science teachers use ICTs a little more proficient than mathematics teachers, except for the more general use of mathematics drill-and-practice programs for remedial purposes. Table 3. Percentage frequencies of mathematic and science teachers who use ICT for different activities Use ICT for the following activities: *

South African mathematics teachers 16.9% a 14.5% a 11.7% b 13.0% b 9.3% 9.0% a

Chilean mathematics teachers 43.3% 40.2% 49.2% 48.9% 21.5% 27.9%

South African science teachers 18.0% b 14.0% a 16.2% c 17.9% b 12.4% 10.3% a

Chilean science teachers 51.1% 40.0% 58.5% 51.2% 25.3% 32.4%

Presenting information, demonstrations or instructions Remedial individual or small group instruction Assisting students in exploratory and inquiry activities Assessing learning through tests or quizzes Effective classroom management Organising, supporting and monitoring team building and collaboration among student * SITES 2006 Mathematics and Science Teacher Questionnaire frequencies for items 14Ab, 14Bb, 14Cb 14Eb, 14Gb and 14Hb, respectively a V = 0.3, medium effect b V = 0.4, medium to large effect c V = 0.5, large effect

4.2 Technical Support of Teachers Sufficient, effective and ongoing technical support and infrastructure is a necessary condition for sustainable integration of ICTs in schools (Chen et al, 2007:online). From Table 4 it becomes evident that in South Africa, the role of a technical coordinator is less defined, but not totally different from the rest of the world. In general, South African teachers are less technically skilled, and rely more on external support than, for example Lithuania, where teachers are mostly responsible for their own technical support. In the majority of schools in Hong Kong, the Moscow Russian Federation, and Singapore, students mostly provide technical support — an indication of their readiness and connectedness for the demands of the 21st century information skills. In South Africa students technically support their schools less than two minutes per week (Pelgrum 2008:91). Table 4. Percentage frequencies of specific categories of people involved in the provision of technical ICT support at schools * Technical support % South Africa Computer coordinator 41% Other ICT staff 24% Teachers 28% * (Pelgrum 2008:92)

70

% Chile 88% 30% 71%

Country with highest % Norway 100% Hong Kong 96% Lithuania 95%


4.3 Teacher Professional Development Two decades of research on the integration of ICT in education shows that although changes are taking place to integrate ICT effectively into teaching and learning, they are not enough to bring about the required change at the required pace (Jamieson-Proctor et al, 2006:512). Factors that hamper the integration of ICT seem to be basically the same as twenty years ago: resistance to change, resources, training and time to integrate ICT effectively (Guru et al, 2005:14). ICTs have the potential to enhance teaching and learning through enriching the curriculum, improving delivery, extending methods of presenting information, and offering new opportunities through the techniques that ICT makes possible. Teachers should also use ICT to manage and reduce their administrative workloads (Becta ICT Research 2004:4). However, few practicing teachers know exactly how to proceed with the integration of ICT into the curriculum. They also do not understand what is actually meant by ICT integration (Jimoyiannis et al, 2007:158). Table 5 compares the need for teacher professional development for the integration of ICTs into teaching and learning in South Africa and Chile. From the secondary analysis it becomes evident that there are significant differences between the insecurities of the teachers of the two developing countries about use ICT for teaching and learning. Preparing of lessons, identifying suitable teaching and learning situations for ICT, using ICT to present information and performing basic Internet searches all relate to competencies of integrating ICT in the curriculum. Sufficient, effective, supportive and ongoing teacher professional development (TPD) is essential for teachers’ integration of ICT into their teaching and learning practices (Chen et al, 2007:online). Professional development should continue throughout a teacher’s life, as it is a tool that creates opportunities for growth and learning, increases effectiveness and decreases isolation (Becta ICT Research 2004:online). Quality leadership is widely acknowledged as imperative for successful schools (Ho 2006:online) Table 5. Percentage frequencies of mathematic and science teachers who indicated that they did not have specific ICT competencies Unable to ... *

South African Chilean South African Chilean mathematics mathematics science science teachers teachers teachers teachers Use word-processing programs for letters, etc. 23.4% b 2.1% 20.9% b 2.0% Send emails with attachments 39.7% b 6.25 38.6% c 4.5% 6.9% 38.0% b 8.8% Use spreadsheets for budgets, administration, etc. 36.7% b Prepare lessons that use ICT 50.7% c 6.9% 50.8% c 5.9% 4.5% 45.5% d 3.9% Identify suitable teaching and learning situations for ICT 46.7% c Search for curriculum resources on the Internet 43.6% d 3.0% 42.8% d 2.6% 13.9% 48.5% b 9.6% Use ICT for learner progress monitoring 50.1% b Use ICT for presentations or explanations 48.7% c 11.2% 48.1% c 7.6% 9.8% 50.4% c 10.7% Install educational software 49.6% c Perform basic Internet searches 49.8% c 8.0% 47.4% c 6.5% * SITES 2006 Mathematics and Science Teacher Questionnaire frequencies for items 21A, 21B. 21E, 21I–21M, and 21O-21P, respectively a V = 0.3, medium effect b V = 0.4, medium to large effect c V = 0.5, large effect d V = 0.6, large effect

Principals are the cornerstone in the promotion of innovative use of ICTs in their schools and should encourage teachers to use ICT in their instructional practices (Ho 2006:online). Principals should pay attention to a wide array of factors that influence teachers’ effective use of ICT, understand the value of ICT in education, and positively motivate teachers to use of ICTs in their classes. To successfully integrate ICT into a school’s education system requires a new way of thinking, as well as actions to create appropriate TPD training for effective and sustainable ICT integration. Principals should exhibit effective managerial and leadership skills to facilitate change (Vallance 2008:290). Table 6 indicates the perceptions of principals regarding the competencies of teachers of the 21st century. Due to recent changes in the South African education system, aspects such as integrated Web-based learning, using new ways of assessment, and being knowledgeable about the pedagogical issues of integrating ICT into teaching and learning are advocated in the first ICT policy document, the e-Education White Paper (Department of Education 2004:1-70).

71

ISBN: 978-972-8924-82-9 © 2009 IADIS

Table 6. Percentage frequencies of knowledge and skills required at schools according to school principals Knowledge and skills required * South Africa Chile Country with lowest Integrated Web-based learning 22.1% a 7.7% Spain 1% Using new ways of assessment 43.4% a 13.7% France 2% Developing real-life assignments for students 30.2% 24.9% Denmark 2% Using computers to monitor student progress 25.2% 13.3% Italy 2% Being knowledgeable about the pedagogical issues of integrating 22.8% a 11.1% Finland 2% ICT into teaching and learning Using subject-specific learning software (tutorials, simulations, etc.) 23.0% 11.5% Norway 2% * SITES 2006 Principal Questionnaire frequencies for items 12A, 12B, 12C, 12E, 12I and 12J, respectively a V = 0.3, medium effect

Principals can implement various strategies and facilitate many channels to assist teachers to increase their skills and knowledge on the effective integration of ICTs. Becta (Becta ICT Research 2004:3-8) suggests that that such channels could be categorised into three categories: teacher-based strategies, schoolbased strategies, and external-based strategies. The South African data indicates that school-based channels are under utilised, and that the technical coordinators at schools could provide more effective support to teachers (Table 7). Table 7. Percentage frequencies for different channels used for teacher development according to technology coordinators at schools Channels used for teacher development * South Africa Chile Country with lowest Informal contacts or communication 50.6% 67.1% Canada 96% ICT coordinator or technical assistant 31.2% c 76.9% Hong Kong 99% In-school courses 36.3% a 67.5% Singapore 98% Training from a teacher who attended a course 45.5% 66.1% Chinese Taipei 92% 80.6% France 98% Observation of and discussions with colleagues 50.9% a Reading of professional journals and similar 25.6% 38.6% Slovak Republic 77% publications * SITES 2006 Principal Questionnaire frequencies for items 11A, 11B, 11C, 11D, 11I and 11J, respectively a V = 0.3, medium effect c V = 0.5, large effect

Teacher-based strategies intend to assist teachers in the integration process. A variety of approaches, building on previously gained skills and competencies, contribute towards effective use of ICT. It is essential to not consider ICT, content and pedagogy in isolation, but the complex relationships with one another as well. Teachers should be skilled in applying pedagogy to the specific content, transform the content through application of ICT, and use ICT to support pedagogical skills (Becta ICT Research 2004:40). Table 8. Percentage frequencies of different types of courses available for teachers’ development, internally and or externally according to the technical coordinators * Available courses for teacher development ** South Africa Chile Introductory courses for general applications (basic word32% 54% processing, spreadsheets, databases, etc.) and Internet use Technical courseware for operating and maintaining computer 19% a 26% systems 26% Advanced courses for applications or standard tools 21% b (advanced word-processing, relational databases, etc.) Advanced courses for Internet (designing web pages, using 16% 24% video conferencing, etc.) Courses on pedagogical issues related to integrating ICT into 15% c 51% teaching and learning Courses in subject-specific training software to attain learning 17% 32% outcomes Courses on multimedia use, for example digital video and / or 13% 22% audio equipment * (Pelgrum 2008:100-101) ** SITES 2006 Principal Questionnaire frequencies for items 12A-12G, respectively

72

Country with Lowest Lithuania 98% Estonia 97% Slovenia 87% Lithuania 86% Lithuania 92% Lithuania 92% Singapore%


The e-Education policy document provides a framework for the collaboration of government and the private sector in the provision of ICTs in education to prepare schools for quality learning and teaching for the 21st century: “Every South African manager, educator, and learner in the general and further education and training bands will be ICT capable (that is, use ICTs confidently and creatively to help develop the skills and knowledge they need as lifelong learners to achieve personal goals and to be full participants in the global community) by 2013” (Department of Education 2004:17). Table 8 lists categories of courses that could assist teachers in realising the strategic goal of the policy document.

5. CONCLUSIONS ICTs have the potential to enhance teaching and learning through enriching the curriculum, improving delivery, extending methods of presenting information, and offering new opportunities through the techniques that ICT makes possible. Despite the difficult teaching and learning conditions in South Africa, ICTs and their role in education cannot be ignored. The rapid development of ICTs internationally has stimulated debate about the roles that these technologies can play in accelerating social development in a developing country such as South Africa. Most of this debate in South Africa tends to centre on the extent to which access to and use of ICT can contribute to reducing the massive inequalities in the population (Department of Education 2006:5). On the one hand, there is the conviction that South Africa will benefit from the development of ICT. Building South Africa’s Information Society would, amongst other things, provide substantial opportunities for creating new jobs in the emerging global information-based economy. It would provide opportunities for improving quality and distribution and for reducing costs of healthcare, would link all educational institutions and provide access to distance education and the world’s best libraries, and open up new markets for trade. Conversely, there is some awareness of the danger that rapid ICT growth may serve to embed inequality and widen the technological gap between those who are connected and those who are not (Daniël et al, 2007:636). Table 9 indicates that more than half of the respondents that participated in the SITES 2006 study were of the opinion that using ICT in teaching and learning had a positive impact on education, especially in terms of 21st century skills. Table 9. Percentage frequencies of Mathematic and Science teachers who indicated that ICT had a positive impact on learner development in their target classes Positive Impact *

South African Chilean Mathematics South African Chilean Science Mathematics Teachers Teachers Science Teachers Teachers Subject matter knowledge 62.1% 93.6% 60.6% 96.6% Learning motivation 63.0% 97.5% 60.0% 97.7% Information-handling skills 58.5% 94.5% 60.5% 97.2% Problem-solving skills 58.6% 87.9% 57.5% 89.2% Self-directed learning skills 55.9% 89.1% 54.8% 87.7% Collaborative skills 58.9% 89.0% 55.8% 86.2% Communication skills 63.2% 90.3% 59.0% 89.2% ICT skills 55.5% 96.1% 58.0% 95.7% Ability to learn at own pace 59.0% 91.3% 58.2% 91.2% Assessment results 57.0% 82.1% 60.9% 43.8% Narrowing the achievement gap 49.4% 77.6% 55.0% 77.3% Narrowing the digital divide 39.9% 46.1% 47.1% 44.6% * SITES 2006 Mathematics and Science Teacher Questionnaire frequencies for items 20A–20K, and 20O, respectively

After considering the results of this secondary analysis Aesop’s fable does not apply well, but it seems if the paradox of Achilles and the tortoise relates more closely to integrating ICT into teaching and learning in South Africa: Achilles (South Africa) is in a footrace with the tortoise and wants to catch up with the tortoise that has a head start. If they would both run at a constant speed, one fast and the other slow, then after some finite time, Achilles will reach the tortoise’s starting point. During this time, the tortoise has also moved on. It will then take Achilles some further time to run that distance, in which time the tortoise will have advanced farther. Thus, whenever South Africa reaches somewhere the Northern hemisphere has already been, South Africa still has farther to go ... The Minister of Education, Naledi Pandor, confirms this position with her statement: “Not yet where we want to be” (Pandor 2007:online).

73

ISBN: 978-972-8924-82-9 © 2009 IADIS

ACKNOWLEDGEMENTS Acknowledgement to: (i) Dr. Suria Ellis from the Statistical Consultation Service, North-West University, Potchefstroom Campus, for statistical support; and (ii) the National Research Foundation for project funding.

REFERENCES Anderson, R. et al, 2008. National Contexts. In Pedagogy and ICT Use in Schools Around the World: Findings from the IEA SITES 2006 Study. Law, N. et al. Hong Kong, Springer and Comparative Education Research Centre. CERC Studies in Comparative Education 23, pp. 37-66. Becta ICT Research, 2004. What the Research Says about ICT and Continuing Professional Development (CPD) for Teachers. Retrieved 8 July 2006, from http://www.becta.org.uk/page_documents/research/wtrs_cpds.pdf. Chen, J. C. et al, 2007. Testing the Whole Teacher Approach to Profesional Development: A study of Ebhancing Early Chilhood Teachers' Technology Proficiency. Retrieved 28 July 2006, from http://ecrp.uiuc.edu/v8n1/chen.html. Daniël, J. et al, 2007. Exploring the Role of ICTs in Addressing Educational needs: Identifying the Myths and Miracles. South African Journal of Higher Education Vol 21, No. 6: pp. 634 - 644. Department of Education, 2004. White Paper on E-Education: Transforming Learning and Teaching through Information and Communication Technologies (ICTs). Pretoria, Government Printing Works. Vol 1: 1-70. Department of Education, 2006. Annual Report 2005/2006. Department of Education. Cape Town, Government Printer. Vol: 161. Department of Education, 2007. National Education Infrastructure Management System. Retrieved 20 July 2007, from www.education.gov.za. Gonzales, P. et al, 2004. Highlights From the Trends in International Mathematics and Science Study (TIMSS) 2003. Retrieved 2 February 2009, from http://www.colorado.edu/physics/phys4810/phys4810_fa08/refs/TIMMS_2003.pdf. Guru, M. et al, 2005. Recapturing Technology for Education. Keeping Tomorrow in Today's Classrooms. New York, Rowman & Littlefield. Ho, J., 2006. Technology Leadership. Retrieved 10 March 2008, from http://www.moe.gov.sg/edumall/rd/litreview/techn_leadership.pdf. Jamieson-Proctor, R. M. et al, 2006. ICT Integration and Teachers' Confidence in Using ICT for Teaching and Learning in Queensland State Schools. Australian Journal of Educational Technology Vol 22, pp. 511-530. Retrieved 16 August 2007 from http://www.ascilite.org.au/ajet/ajet22/jamieson-proctor.html. Jimoyiannis, A. et al, 2007. Teacher Development. Teacher Development Vol 11, pp. 149-173. Retrieved 14 August 2007 from http://www.ecedu.upatras.gr/servicces/people/files_publications/TD_JimyiannisKomis.pdf. Law, N. et al, 2008a. Study Design and Methodology. In Pedagogy and ICT Use in Schools Around the World: Findings from the IEA SITES 2006 Study. Law, N. et al. Hong Kong, Springer and Comparative Education Research Centre, The University of Hong Kong, pp. 16-37. Law, N. et al Eds. 2008b. Pedagogy and ICT Use in Schools around the World: Findings from the IEA SITES 2006 Study. Hong Kong, Springer and Comparative Education Research Centre, Universities of Hong Kong. pp. 1-296. Onwu, G. Ed. 1998. Teaching large classes. African science and technology education into the new millennium: practice, policy and priorities. Johannesburg, Juta. pp. 119-132. Pandor, N., 2007, 9 January 2007. E-learning in South Africa. Address by the Minister of Education, Naledi Pandor MP, at the World Ministerial Seminar on Technology in Education Retrieved 14 February 2007, from http://www.education.gov.za/dynamic/dynamic.aspx?pageid=306&id=2471. Pelgrum, W., 2008. School Practices and Conditions for Pedagogy and ICT. In Pedagogy and ICT Use in Schools Around the World: Findings from the IEA SITES 2006 Study. Law, N. et al. Hong Kong, Springer and Comparative Education Research Centre, The University of Hong Kong, pp. 67-120. Pelgrum, W. et al, 2008. Introduction to SITES 2006. In Pedagogy and ICT Use in Schools Around the World: Findings from the IEA SITES 2006 Study. Law, N. et al. Hong Kong, Springer and Comparative Education Research Centre, The University of Hong Kong, pp. 1-11. South Africa.info, 2006, December 2006. South Africa: Fast Facts. Retrieved 14 February 2007, from http://www.southafrica.info/ess_info/sa_glance/facts.htm. Vallance, M., 2008. Beyond Policy: Strategic Actions to support ICT Integration in Japanese Schools. Australian Journal of Educational Technology Vol 24, pp. 275-293. Retrieved 27 August 2008 from http://www.ascilite.org.au/ajet/ajet24/vallance.pdf.

74


ANTI-SOCIAL NETWORKING SITES: WEB 2.0 AND SOCIAL EXCLUSION Denise Leahy, Ultan Ó Broin Trinity College Dublin, Ireland

ABSTRACT The European Union (EU) promotes the concept of eInclusion as part of the i2010 initiative (European Commission, 2005). This includes areas such as eAccessibility, Digital Literacy and eGovernment – all to “improve people’s quality of life”. Internet-based economic opportunity and political engagement are also part of the socially driven Web 2.0 concepts of participation and collaboration. Increasingly, Web 2.0 technologies are adopted by enterprises to integrate with the collective intelligence of the community at large, for example using social networking sites for sales opportunities (McKinsey, 2007). If accessibility is not built into these systems, people with disabilities may be excluded from social interaction, political organization, economic, and other opportunities. Despite the widely accepted claims about Web 2.0’s inclusiveness based on participative patterns of usage (O’Reilly, 2004), (Madden and Fox, 2006), is the lack of accessibility support within Web 2.0 technology itself actually creating social exclusion? This research looks at Web 2.0 accessibility challenges by examining the social networking site experiences of a group of users with visual impairments compared with a group of sighted users. KEYWORDS Accessibility, social networking, Web 2.0

1. INTRODUCTION The Internet now plays an important part in the lives of many. However, those without access to the Internet may be missing opportunities to participate fully in the Information Society. Research by Nielsen (2006a) and the Worldwide Web Consortium (1999) show how disability can negatively impact Internet participation. Concern about this impact on society is reflected by the emergence of the policy concept of eInclusion and the following of web accessibility guidelines to enable equal opportunity through the Internet - “It is essential that the Web be accessible in order to provide equal access and equal opportunity to people with disabilities” (Henry, 2007). The eEurope vision is of an Information Society that releases human and economic potential to improve productivity and the quality of life for the citizens of Europe (ETSI, 2003). The European Commission cites the main causes of social exclusion as unemployment and the lack of access to resources and training (European Commission, 2002). As people increasingly adopt what is called “Web 2.0” to interact with others, is this creating a new kind of digital social exclusion of persons with a disability? This research examines the use of social networking sites (Facebook, MySpace, Bebo, etc) by Irish users as a good indicator of social inclusion, as such sites “allow individuals to present themselves, articulate their social networks, and establish or maintain connections with others” (Ellison, Steinfield, and Lampe, 2007).

75

ISBN: 978-972-8924-82-9 © 2009 IADIS

2. HOW IS WEB 2.0 BEING USED? 2.1 Social Networking Social networking sites are a people based concept that can be work-related, (such as LinkedIn), romantically oriented (Friendster), or shared interest and social relationships (Bebo, MySpace, Facebook and others). Facebook had 123.9 million unique visitors in May 2008, MySpace 114.6 million, and Bebo 25.1 million (Schonfeld, 2008), and the uptake is increasing globally. The use of these social networking sites represents the very essence of user collaboration and participation on a mass scale. These sites allow users to post and share content, links, images, video, music, join and create online groups of common interest, engage in online debate and other exchanges, mail and instant message each other. Social inclusion, or eInclusion, is an area of increasing interest for academics, practitioners, and policy makers. The centrality of Internet technology to everyday lives and its potential to provide opportunity to alleviate disadvantage is recognized. For example the eEurope “Information Society” (European Council, 2005) aims to improve productivity and quality of life for the citizens of Europe (ETSI, 2003) through modern online public services; including e-government, e-learning, and e-health for citizens working in an ebusiness environment. However, not everybody has access to online environments and there are different reasons for the adoption of technology across different groups of users in the community (Venkatesh, et al, 2003). People with disability face barriers when using websites and services simply because the online technology and content do not support how they use the web. To address these issues there are a wide variety of guidelines and regulations, best known of which is the voluntary guidelines of the Web Accessibility Initiative (WAI) (1999), supported by local, national, and international legislation and aspirations. Notwithstanding the European Telecommunications Standards Institute (2003) claim that Internet technology is an opportunity “for disabled groups and other vulnerable groups to gain equality of access to participation in society not just as another means by which they are to be disenfranchised or excluded”, Shawn Henry (Thatcher, et al, 2006) argues that although web accessibility is essential for equal opportunity we constantly need to work to ensure “the Web be accessible in order to provide equal access and equal opportunity to people with disabilities”.

2.2 User Created Content The OECD’s (2007) report “Participative Web and User-Created Content: Web 2.0, Wikis, and Social Networking” also recognizes the social, cultural economic opportunities and impacts of participative content, referred to as user-created content (UCC). This they define as: “Content made publicly available over the Internet which reflects a certain amount of creative effort”, and which is created outside professional routines and practices. It is this creation of content that has major social implications because it has “altered the economics of information production, increased the democratisation of media production, and led to changes in the nature of communications and social relationships.” Such content has great potential to increase user participation and diversity and the volume is rapidly expanding, with a minimum of 130 million pieces of content under Creative Commons licences alone by mid-2008 (Creative Commons, 2008). Web 2.0 is perceived as mainly a social phenomenon, changing the patterns of “who communicates with whom, under what conditions, and at whose discretion” (Benkler, 2006). Tapscott and Williams (2006) state “the new web is fundamentally different in both its architecture and its applications… Whether people are creating, sharing, or socializing, the new Web is principally about participating rather than about passively receiving information”. Lessig’s work (2005) reveals how participatory web empowers a participatory culture in society An AbilityNet (2008) survey found that the most popular social networking websites on the Internet today are “either difficult or impossible for disabled people to use – in many cases a user is not even able to register with the website.” Zajicek (2007) defines accessibility in a way of particular interested to participation on the web - “A community web site is accessible if it includes the user in its group and the user wants to be included. If you are excluded from a service, then it is not accessible to you”. Raman (2009) says, “A

76


significant portion of our social interaction increasingly happens via the Web”. So, who is using Web 2.0 and is the lack of accessibility within Web 2.0 technology creating social exclusion?

3. RESULTS OF THE SURVEY The survey was distributed to visually impaired users using announcements sent to the Irish-based Visually Impaired Computer Society (VICS) forum (http://vicsireland.org/), as well as to other private and public groups working in the area of visual impairment. The research was broad, but only the parts that relate use of social networking are discussed in this paper. Despite user pre-testing and checking against the Web Content Accessibility Guidelines 1.0 (1999), some users with visual impairments experienced difficulty and did not complete all survey questions. This was addressed by relabeling some options, rewriting instructions, and adding details of switching into Forms mode in JAWS. The problem was caused by the different assistive technologies’ handling of web forms and the varying user expertise with the same assistive technology (Thatcher et al, 2006). This needs to be borne in mind for all researchers in the area. Respondents were asked if their usage of these web sites was passive - such as merely having an account, or activities like viewing or reading; or more active - such as editing, contribution, posting. The research literature indicates most usage is largely passive (Nielsen, 2006a). Respondents were asked about the challenges to their usage of the sites and services and the seriousness of these challenges on a number of areas (age, social circle, privacy fears, content mistrust, and so on), and then to give their opinion on the best approach to achieving the desired accessibility. Open-ended questions were asked and the respondents were invited to comment on how users who have visual impairments could best influence web site development or others users who create content or relationships to deliver an accessible user experience for all (for example, by providing feedback, leveraging legal, political, social processes, and so on).

3.1 Profile of the Respondents 20 sighted users and 29 users with visual impairments completed the survey. More than two thirds of the respondents with visual impairments were completely blind, with low vision making up the second most common visual impairment. JAWS, Windows-Eyes and other screen readers were reported as the most common assistive technology used (79.3%). Most respondents were in the 25-35 age range; 35% of those with visual impairments and 60% of sighted respondents were in this group. Most of the other sighted respondents were aged between 35 and 45, while the remaining 65 % of respondents with visual impairments were evenly spread across all age groups.

3.2 The Use of Social Networks For sighted respondents, there was almost universal usage of Wikipedia (94.7%), followed by YouTube with the next highest usage (89.5%), followed by Amazon (73.7%), and then social networking sites (63.2%) and EBay (63.2%). For visually impaired respondents, the pattern is different. Although Wikipedia is the most widely used site or service (75%), it was closely followed by Amazon (71.4%), while no other category of site or service that could be considered “Web 2.0” made it past the 50% mark. Sighted respondents expressed very strong or strong reasons for using social networking sites like Facebook, Bebo and MySpace and such services as: being part of social groups of common interest (52.9%); obtaining opinions on goods and services by real users (strong and very strong reasons were both 29.4%); finding out information about jobs and career development (58.8%); wanting to find out more information (55.6%); and making new friends or linking up with new ones (44.4%). Visually impaired respondents showed less interest in using such technology to make new friends and link up with old ones (34.8%), and a conflicting equally strong and neutral reason for being part of social groups of common interest (26.1%). Using the technology for career development was a very weak interest (30.4%). Using web sites and services to obtain opinions on goods and services by real people and finding out more information was recorded as a

77

ISBN: 978-972-8924-82-9 © 2009 IADIS

strong reason (50%) and very strong reason respectively (56.5%). Making input to debates and reading the opinions and recommendations of others were also strong (40.9% and 45.5%). We can conclude that although all respondents were interested in using Web 2.0 for individual reasons (shopping, finding out information, and so on), with visually impaired users this is less to do with social networking or employment networking, which may have implications for inclusion and building social capital across the community as a whole.

3.3 Challenges to Information Sharing and Collaboration The major issue reported by most respondents was the use of an inaccessible Captcha (Completely Automated Public Turing test to tell Computers and Humans Apart) at sign-up time (94.7%). Other major challenges included: the inability of screen readers to detect changes on dynamic page (70%); badly designed online forms for data entry (70%); lack of ability to determine content of visual elements (65%); and no ability to control interactive elements such as audio and video players (63.2%). These, and the remaining challenges are all well known by accessibility practitioners and users of assistive technology alike, and widely acknowledged by accessibility guidelines as serious areas for redress, but yet they remain present. Of the top five accessibility challenges, the most serious, the use of inaccessible Captchas (Figure 1) precludes any further involvement with social networking sites (unless visually impaired users obtain help from another person to proceed.) The next most serious challenge, badly designed forms, prevents users from collaborating and participating, as online forms are widely used to enter data, submit comments and feedback; while the inability of assistive technology to determine dynamic changes indicates that the widely used technologies (notably Asynchronous JavaScript and XML [AJAX]) used for Web 2.0’s rich user experience is a problem. Visually-impaired users also experienced problems with the highly graphical, multimedia, and user-created content that make up a very significant percentage of Web 2.0 content, represented by the fourth and fifth most serious challenges recorded.

Figure 1. Inaccessible Captcha example from Bebo.com

The survey also explored the other concerns users had when using Web 2.0 sites and services. For both groups, major issues identified were privacy fears and content mistrust. However, for visually impaired respondents, the major challenge was the lack of accessibility support in the technology itself (80%).

78


These respondents were asked to rank the seriousness of the different types of accessibility issue they experienced using the sites and services mentioned. The details of these challenges are shown in Table 1 “Accessibility challenges of users with visual impairments to using Web 2.0 sites and services.” Table 1. Accessibility challenges of users with visual impairments to using Web 2.0 sites and services (ranked by number of respondents and percentage of total) Major Issue Use of inaccessible Captcha on signup Badly designed online forms for data entry Inability of screen readers to detect changes on dynamic page No ability to determine visual content with text (e.g. no Caption, Title or Alternative on images) No ability to control interactive elements such as audio and video players Videos with no soundtrack or text transcript alternative Use of specific colours to indicate functionality Requirements to add plug-ins before the content can be accessed. Complex tables used for illogical layout Inability to expand links or show hidden text Complicated, wrongly marked up data tables that confuse screen readers Unclear text-speak language and abbreviations in content No ability to Navigate Colour-combinations on text or backgrounds No keyboard support on keys, links, hot-keys, shortcut keys etc. Inability to control text size on content

Slight Issue

Neutral Issue

Not an Issue

18 (94.7%)

0 (0%)

1 (5.3%)

0 (0%)

14 (70%)

5 (25%)

0 (0%)

1 (5%)

14 (70%)

5 (25%)

0 (0%)

1 (5%)

13 (65%)

4 (20%)

1 (5%)

2 (10%)

12 (63.2%)

2 (10.5%)

4 (21.1%)

1 (5.3%)

9 (45%)

8 (40%)

2 (10%)

1 (5%)

7 (35%)

5 (25%)

6 (33.3%)

2 (10%)

7 (35%)

4 (20%)

7 (35%)

2 (10%)

6 (31.6%)

3 (15.8%)

7 (36.8%)

3 (15.8%)

6 (31.6%)

4 (21.1%)

5 (26.3%)

4 (21.1%)

6 (31.6%)

8 (42.1%)

4 (21.1%)

1 (5.3%)

6 (30%)

5 (25%)

5 (25%)

4 (20%)

6 (30%)

5 (25%)

3 (15%)

6 (30%)

4 (20%)

1 (5%)

4 (20%)

11 (55%)

4 (20%)

6 (30%)

4 (20%)

6 (30%)

2 (11.1%)

1 (5.6%)

5 (27.8%)

10 (55.6%)

3.4 Active and Passive Use of Web 2.0 The figures for passive usage of the web sites and services show that for sighted respondents the strongest usage is looking up information on Wikipedia followed by reading comments feedback and ratings and having accounts on social or employment related sites. Visually impaired respondents also expressed a strong preference for Wikipedia, but with a lower score for reading comments, feedback, and ratings. However, having accounts on social or employment related sites recorded a much lower score than sighted users, though the figure for reading blogs is higher (60%). Here, we can also conclude there are implications for social inclusion given the potential of social networking sites for community or political organization and economic opportunity. In terms of active participation, we can expect lower figures than for passive (Nielsen, 2006b). From the survey, we find that for sighted respondents, posting images on photo sharing sites was the top activity (77.8%), followed by submitting feedback, comments, reviews, or ratings to a site (55.6%) and posting blog articles (50%). For visually impaired respondents, the top activities were submitting feedback, comments, reviews or ratings to a site (55%), posting a blog article (40%), and editing a Wikipedia article (36%). Not

79

ISBN: 978-972-8924-82-9 © 2009 IADIS

surprisingly, posting images to photo sharing sites records a much lower score (5%). “None of these activities” recorded 32%. The implications for social inclusion here are clearly in line with the AbilityNet report (2008), but also indicate that the concerns expressed over the ability of existing guidelines to deliver an accessible web are valid (Kelly, et al, 2007), (Burnett, 2006).

4. HOW CAN THIS BE ADDRESSED? All respondents were asked how to best achieve accessibility in Web 2.0 sites and services. Visually impaired respondents were very specific and discussed areas of education, lobbying, and the technical issues, which should be addressed. The most common sentiment was to “lobby government agencies, European and UN agencies. Educate web developers. Support all regulatory organizations such as W3C, etc.” Sighted respondents identified the provision of development tools with built in accessibility for content creation, the following of coding standards, and information for content developers and beta testing as very important. The employment of usability experts with accessibility expertise and offering free screen readers was seen as somewhat important, with a neutral opinion on the policing of sites for non-accessible content. Visually impaired respondents were unanimous in stating the importance of following the widely accepted web accessibility guidelines, the provision of development tools that build in accessibility support when content is created, accessibility information for content developers, beta testing before rollout and employment of usability experts with accessibility expertise. The provision of free screen readers was recorded as of “neutral” importance and the policing of non-accessible content was very important. All respondents said that coding standards and the accessibility guidelines were important. Respondents were also asked about their agreement with statements about the strategic direction of accessibility. Sighted respondents somewhat agreed that web accessibility is becoming more about the flexibility of assistive technology with alternative versions of web sites becoming the norm, and completely agreed that user generated content was always likely to offer poor accessibility. Respondents with visual impairments completely agreed with the statement about the flexibility of assistive technology. However, they completely disagreed that alternative accessible versions of web sites will become the norm, and somewhat agreed with the likelihood that user-generated content would always offer poor accessibility. From this we can see recognition of the challenge of user-generated content in terms of accessibility, but also the recognition that web accessibility guidelines alone are not providing accessibility. Visually impaired respondents reflected their “overall” usability and wider stakeholder participation concerns by indicating that “alternative” sites and services are not acceptable (Horton, 2005). Comments about accessibility of current web sites included: • “As a person, who is totally blind, I cannot, at this time, sign myself on a My Space, Plaxo or LinkedIn account independently because I get stuck in all of the boxes. So I have to have someone help me, which is very disheartening since I need to be a part of these networking sites for my profession. I’m feeling that, in my career, I am not able to keep up with my colleagues, which is a little scary. • “So-called accessible versions of websites have been disastrous up to now, and are less of an answer than properly structured web sites. Tesco in the UK came adrift with this one. The ‘accessible’ site was never kept up to date, didn’t have special offers, and anyway formed a kind of ‘ghetto’ for disabled people. Special provision for disabled people could cause resentment from other users and those who have to build the services on the sites, not to mention the disabled people themselves.” • “When major players such as Yahoo and MySpace won’t even reply to e-mail about their inaccessible Captcha systems, there is still quite a long way to go in breaking down barriers. On the other hand, Google, Twitter and others have been amenable to changing their systems. Technically, however, audio Captcha are going to be easy to crack with speech recognition, so seems the future is uncertain here.” The technical issues identified by the users were: • “Need a way of making screen readers tell us how we can use all the new types of link and control building into coding so they can speak them like tutor messages, and need screen readers to automatically read new text on existing pages.”

80


• •

“I’ve heard of WYSIWYG web development tools which prompt for ALT tags... but we all know most people just ignore those!” “The increasing prevalence of AJAX poses a great problem to the goal of a more accessible Internet. It is important that the tools used to generate web content incorporate accessibility by default. These days anybody in the world can create a website. We can’t expect everybody to have an understanding of the requirements needed to create accessible content. Therefore, it must be made as easy as possible by building it into the tools used.”

5. CONCLUSION The survey results and comments support the wide recognition of the importance of the collaborative and participative features of Web 2.0 sites and services. In their use of the Web 2.0 technology, respondents with visual impairments showed similar patterns to sighted respondents in their looking up of information, reading blogs and comments, feedback and ratings on sites like Amazon and eBay. However, there is a marked difference in usage when it comes to using social networking sites like Facebook, MySpace LinkedIn and Bebo, which have come to define the very collaborative and participative nature of Web 2.0 and are very widely used (Schonfeld 2008). These findings have serious implications for social inclusion. Visually impaired respondents recognized the learning, social, lobbying and employment potential of such sites, but in using such sites they are presented with some very serious accessibility challenges. As a result the respondents mentioned being “disheartened”, “unable to keep up with colleagues” and “banished” from these sites. When questioned on the challenges presented to using the sites and services mentioned in the survey, it is shown that accessibility is by far the greatest challenge to the visually impaired user. The top five accessibility challenges alone identified by respondents are all synonymous with the technology used to provide Web 2.0’s rich user experience. Although there are other issues of concern to all; issues like privacy, content trust worthiness and issues like age, occupation, and social circle are similar between visually impaired and sighted users, it is clear that social networking sites are to a large extent inaccessible and therefore the visually impaired user must be considered as being socially excluded. This is contrary to not only well-known accessibility guidelines such as the WCAG but also the claims of Web 2.0 thought leaders, and the aspirations of the EU and other public policy bodies. This exclusion has serious implications on a community and individual level, as visually impaired users can be excluded from social interactions, and also political processes and economic opportunity. What is of major concern about these findings is that none of the top ten accessibility challenges identified is an unknown quantity from the accessibility guidelines and usability viewpoint. Even the more “recent” accessibility issues relating to key Web 2.0 technologies such as AJAX are currently being addressed through WAI-ARIA (Worldwide Web Consortium, 2008) or other guidelines (Gibson, 2006). What is lacking is the willingness to apply these guidelines. Given that the more straightforward issues like tables, images or links were not addressed in our research it is hard to see how more complex, newer, ones will be. This is an area of great potential for further research, using more extensive survey techniques, larger respondent pools, and exploring impacts of factors such as other disabilities (age, for example) and other reasons for uptake of technology by respondents. If lessons are not heeded by policy-makers and technology innovators, the likelihood for dealing with accessibility challenges in other important and emerging, and sometimes Web 2.0 related computing platforms such as mobile (Hartley, 2008), (Abrahams, 2008) is not very promising, thus increasing social exclusion further.

81

ISBN: 978-972-8924-82-9 © 2009 IADIS

REFERENCES AbilityNet. 2008, 19 August 2008, State of the eNation Reports: Social networking sites lock out disabled users. Online, Available from: http://www.abilitynet.org.uk/enation85 (accessed 19th February 2009) Abrahams, P. 2008. Why not make the iPhone more Accessible? Available from: http://www.itanalysis.com/business/change/content.php?cid=10678 (accessed 19th Feb 2009) Burnett R. 2006, Disability Discrimination and Internet Accessibility, Accountancy Books: UK, ISBN: 9781841524122 Creative Commons, 2008, Metrics, Available from: http://wiki.creativecommons.org/Metrics (accessed 18th February 2009) Ellison B, Steinfield C, and Lampe, C., 2007, The Benefits of Facebook “Friends:” Social Capital and College Students’ Use of Online Social Networking Sites in Journal of Computer-Mediated Communication Vol 12, 2007, pp 11431168. European Commission, 2002, Delivering eAccessibility: Improving disabled people’s access to the Knowledge Based Society. Available from: http://www.epractice.eu/document/4074 (accessed 19th February 2009) European Commission, 2005, i2010 - A European Information Society for growth and employment, Online, Available from: http://ec.europa.eu/information_society/eeurope/i2010/index_en.htm, (accessed 18th February 2009) ETSI, European Telecommunications Standards Institute, 2003, Human Factors (HF); Mutlimodal interaction, communication, and navigation guidelines, ETSI Guide 202 191 v1.1.1 (2003-08) Gibson B, 2006. JavaScript and AJAX Accessibility, Available from: http:// www03.ibm.com/able/dwnlds/AJAX_Accessibility.pdf (accessed 18th February 2009) Hartley S, 2008, Global Mobile Market Outlook: 2008-2013. Available from: http://www.ovum.com/go/content/s,75685, (accessed 19th February 2009) Henry, L S. 2007, Just Ask: Integrating Accessibility Throughout Design, Madison, WI: ET\Lawton, 2007. ISBN 9781430319528 www.uiAccess.com/JustAsk/ (accessed 19th February 2009) Horton S. 2005, Access by Design: A Guide to Universal Usability for Web Designers. New Riders Press: UK Kelly B, Sloan D, Brown S, Seale J, Petrie H, Lauke P, Ball S. 2007. Accessibility, Policies and Process, Technical paper, 16th International Worldwide Web Conference Lessig, L. 2005. Free Culture: The Nature and Future of Creativity. New York: USA. Madden M and Fox S. 2006 Pew Internet Project. Riding the Waves of “Web 2.0”: More than a Buzzword but still not easily defined. Available from: http://www.pewinternet.org/PPF/r/189/report_display.asp (accessed 19th February 2009) McKinsey Quarterly, 2007, How Businesses are using Web 2.0: A McKinsey Global Survey. Available from: http://www.mckinseyquarterly.com/information_technology/management/how_businesses_are_using_web_20_a_mc kinsey_global_survey_1913 March 2007 (accessed 19th February 2009) Nielsen, J. 2006a, Digital Divide: The Three Stages. Available from: http://www.useit.com/alertbox/digital-divide.html (accessed 19th Feb 2009) Nielsen J. 2006b, Participation Inequality: Encouraging More Users to Contribute. Available from: http://www.useit.com/alertbox/participation_inequality.html (accessed 19th February 2009) O’Reilly T. 2005. What Is Web 2.0: Design Patterns and Business Models for the Next Generation of Software, Available from: http://www.oreillynet.com/ (accessed 19th February 2009) OECD, 2007, Participative Web and User-Created Content: Web 2.0, Wikis, and Social Networking, OECD: Paris. Raman T V, “Toward 2w, Beyond Web 2.0”, Communications of the ACM, February 2009, vol. 52, no. 2 Schonfeld, E. 2008, Facebook Blows Past MySpace in Global Visitors for May. June 20, 2008. Available from: http://www.techcrunch.com/2008/06/20/facebook-blows-past-myspace-in-globalvisitors-for-may/ (accessed 19th February 2009) Thatcher J, Burks M R, Henry S L, Kirkpatrick A, Lauke P, Lawson B, Regan B, Rutter R, Urban M, Waddell C. 2006. Web Accessibility: Web Standards and Regulatory Compliance, Friendof, New York, ISBN: 1590596382 Venkatesh V, Morris M, Davis G, and Davis D. 2003 User acceptance of information technology: Toward a unified view. MIS Quarterly, 2003, Volume 27 Issue 3. Worldwide Web Consortium (W3C), 1999, Web Content Accessibility Guidelines 1.0, Available from: http://www.w3.org/TR/WCAG10/#priorities (accessed 19th February 2009) Worldwide Web Consortium (W3C), 2008, WAI-ARIA Primer: W3C Working Draft 4 February 2008. Zajicek M. 2007, Web 2.0 Hype or Happiness, Proceedings of the 2007 international cross-disciplinary conference on Web accessibility (W4A), Available online from: http://portal.acm.org/citation.cfm?id=1243441.1243453 (accessed 19th February 2009)

82


UNDERSTANDING HUMAN STATES CONDITION BASED ON PHYSIOLOGICAL AND PSYCHOLOGICAL INFORMATION Santoso Handri*, ** Top-Runner Incubation Center for Academia-Industry Fusion, Nagaoka University of Technology * Kamitomiokamachi 1603-1, Nagaoka, Niigata 9402188 JAPAN* Multimedia Nusantara University** Plaza Summarecon, Boulevard Blok M5 Nomor 3, Gading Serpong, Tangerang, Indonesia**

Shusaku Nomura* Top-Runner Incubation Center for Academia-Industry Fusion, Nagaoka University of Technology * Kamitomiokamachi 1603-1, Nagaoka, Niigata 9402188 JAPAN*

Kazuo Nakamura* Management and Information Systems Science, Nagaoka University of Technology* Kamitomiokamachi 1603-1, Nagaoka, Niigata 9402188 JAPAN*

ABSTRACT The impact of extensively usage visual display terminal (VDT) on automatic nervous systems are observed and analyzed using a profile of mood states (POMS) and physiological information, i.e., electroencephalogram (EEG), heart rate variability (HRV) and thermograph. Two experiments, which have different intermittent schedules, were conducted using the Kraepelin psychodiagnostic test. The purpose is to determine human mental states towards the given task. Based on the subject’s expression through POMS questionnaire conducted before and after experiments and measured physiological signal, a network model for evaluating human states was proposed in this study. Hidden Markov model (HMM) and neural network are employed to differentiate human states and mood condition based on the experiments conducted. The effectiveness of the proposed method to determine human states based on the given task is then investigated in this study. KEYWORDS Physiological, psychological, human states, HMM, neural network.

1. INTRODUCTION With recent extensively usage personal computer (PC) in any activities might cause harmful effect related to the mental health problem at an adequate level, especially to the works which highly depending on visual display terminal (VDT). According to an official statistical research of Japanese government on the workers engaged in the deskwork with visual display terminal (VDT), about 80% of them complain about their somatic symptom caused by deskwork such as low vision, eye strain, stiffness in the shoulder and neck, nevertheless 62% of them said not feel at all or not feel explicitly for their mental stresses (Health, Labor and Welfare Ministry of Japan, 2004). International Labor Organization (ILO) reported different impacts on the societies, e.g., the social cost of mental health problem in European Union was estimated to be on average 3 to 4 % of GNP (Gabriel and Liimatainen, 2000). There were several works studying related health problems caused by mental workload. However, only few works have observed directly the human physiological indication responding mental workload. Zhang et al., (2008) reported about estimating visual display terminal (VDT) mental fatigue using multichannel linear descriptors and kernel-principal component analysis-hidden Markov model (KPCA-HMM). Mizuno et al., (2007) examined the relationship between the job satisfaction and work-family balance of Japanese nurses.

83

ISBN: 978-972-8924-82-9 © 2009 IADIS

Healey and Picard (2005) have evaluated mental stress in real-world driving tasks using physiological sensors. They have introduced several information processing techniques in pursuing to evaluate human mental states condition. This study attempts to evaluate human states based on given task by understanding physiological signals generated by human body and responding POMS questionnaire. Physiological sensors are a useful metric for providing feedback about a worker’s state because they can be collected continuously and without interfering with the worker task performance, while the POMS questionnaire results might be relevant to human sensitivity or impression towards the given task. Thus this research proposed a method to evaluate relation between physiological signal and POMS scores based on the given task

2. THE EXPERIMENT PROCEDURE 2.1 Subjects Nine male students between 21 - 23 years olds voluntary participated in the experiment. Personal data were acquired with a standardized interview before recording physiological information. They did not have any health problem during the experimental period, and that they were not being under any medication. Smoking and hard exercise were also prohibited before the experiment. None of them reported on any cardiovascular disease or neurological disorders in the past. This experiment was conducted in conformity with the Helsinki Declaration. All subjects were well informed regarding the purpose and contents of the experiment, and informed consent was obtained before their participation. They were also informed that they had the right to renounce their participation anytime.

2.2. Experimental Procedure Participants were comfortably seated facing a LCD display at about 50 cm. A simple calculation task with a laptop computer was taken as a mental workload in this experiment. It is a simple addition of two doubledigit integers. These integers were repeatedly presented on the laptop monitor every 3.0 sec with changing figures. The subjects were instructed to input the answer of the addition by keyboard as fast and correct as possible. Such a simple calculation task is quite similar to so-called the Kraepelin psychodiagnostic test and which has frequently been introduced for researches investigating physiological responses induced by mental stress. Moreover, the task has typical features of mental workload in our daily life such as routine, simple, boring, and unlimited. Thus, the result of the experiment is expected to be a practical model of mental workload in our daily life. The experiment consisted of 2 schedules of intermittent calculation tasks: (1) subjects were instructed to conduct 2 sets of 18 min of the calculation tasks with 9 min of an intermediate break (experiment A) and (2) to conduct 6 sets of 6 min of the tasks with 3 min of break between the tasks (experiment B). It should be note that the total duration of the calculation task and break was 54 minute in total for each participant doing each experiment as shown in Figure 1. In addition, the subjects were required to fill a profile of mood states (POMS) before and after each experiment. POMS is a measure of six identified mood factors, i.e., tensionanxiety (TA), depression-dejection (D), anger-hostility (A-H), vigor (V), fatigue (F), and confusion (C); it is commonly used for psycho-physiological studies. Three types of physiological sensors were used during the experiment: electrocardiogram (ECG); electroencephalogram (EEG); and thermograph. The data were recorded using commercial BioPAC MP150 systems for ECG and EEG signal, while the thermograph that uses an infrared camera, measured reveals temperature variations on the surface of the face.

84


Figure1. The experiment procedure

3. PCA AND HMM APPROACH FOR ESTIMATING HUMAN STATES 3.1. Reducing the Dimensions of Features Vector Using PCA PCA is a statistical technique whose purpose is to condensate the information of a large set of correlated variables into a few principal components while not throwing overboard the variability present in the data set (Castells et. al, 2007). The principal components are derived as a linear combination of the variables of the data sets; with weights chosen so that the principal components become mutually uncorrelated. Each component contains new information about the data set, and is ordered so that the first few components account for most of the variability. The objective can be achieved by choosing to analyze only the first few factor principal components. The number of principal component is, ideally, known prior to the analysis. In this study, the first 12 principal components are determined by examining the proportion of total variance over 90% explained by principal component.

3.2. Evaluation of Pattern of physiological data using Hidden Markov Model (HMM) The main objective of this section is to develop models and techniques which can apply in real time to track physiological signal and make inferences about the level of arousal of a subject. We envision this study being a useful building block that can be integrated into a computer that uses this information to adapt itself to the needs of the user. This more ambitious idea goes beyond the present scope of this study but is a future research topic in this area. Human physiology behaves like a complex dynamical system in which several factors, both internal and external, shape the outcome. In approximating such system, we are interested in modeling its dynamical nature and given that knowledge of all the independent variables that affect the system is limited. We want to approach the problem in a stochastic framework that will help us model the uncertainty and variability that arise over-time. A class of models that has received much attention in the research community over past years to model complex dynamic phenomena of a stochastic nature is the class of Hidden Markov Models (HMM). HMMs have been widely used for modeling speech and gesture, and are currently an important block of speech recognition systems. Motivated by their flexibility in modeling a wide class of problem, we decided to study the feasibility of using HMMs to model physiological pattern that are believed to correlated with different affective states. A HMM is a stochastic state machine, characterized by the following parameter set: (1) where A is the matrix of the state-transition probabilities, B is the observation probability, and π is the initial state probability. The observation of a HMM are continuous signal representations, called feature vector, modeled by a Gaussian probability density function of the form:

85

ISBN: 978-972-8924-82-9 © 2009 IADIS

(2) where is the observation vector at time t, is the mean vector, and is the covariance matrix at state j. of the model using The estimated parameters are obtained by performing likelihood maximization an iterative procedure such as Baum-Welch method (Rabiner, 1989). The physiological signal generated by human body might have strong correlation with accumulation of human states. The accumulation of human states may then be seen as one of HMM problems. Thus, in this study, the HMM network are employed to estimate accumulation human states based on the observed physiological information. The first stage is to build a system which adapting with the given data by train the from the HMM network. The purpose of HMM training is estimating the model parameters set observation sequences data O. The HMM parameter estimation is carried out by Baum-Welch method which similar with expectation-maximization algorithms. The number of hidden states of the HMM is assumed having three states, i.e., high, medium, and low states. This categorization of hidden states is performed by clustering method, i.e., competitive learning algorithms based on the physiological data representative. This study only considered one Gaussian density function per state. After training, the HMM is then employed to evaluate probabilities of human states condition. PCA

HMM

Feature Extraction

Neural Network

M

H

L

Experiment A :TTBTTBTTB …

T= Task B= Break

Experiment B :TTTTTTBBB …

Figure 2. The conceptual mechanism for understanding human states condition

3.3. Neural Network Neural Network with a back propagation learning algorithm is well known as a supervised classifier method and suitable for building adaptive pattern recognition system (Duda, Hart, and Stork, 2000). Mapping function of a neuron in a network can be written as

(3) where y is the output, fa is activation function, wi is weight of input xi , b is a bias term and N is total input. In order to determine the mapping function, first, the network needs to be trained by using sample data. Learning via back-propagation involves the presentation of pairs of input and output vectors. Among several activation functions of Neural Networks, this study uses sigmoid function as activation function for hidden layers. The advantages of sigmoid functions are easier to train than threshold units, because of better smoothing function in specific range input-output and have upper-lower bound. With sigmoid function, a

86

M A C H I N E


small change in the weights will usually produce a change in the outputs, which makes it possible to tell whether that change in the weights is good or not. The selection of an activation function of the output units should suit with the distribution of the target values. In this study, the identity or linear activation functions are employed. In this work, Neural Network Toolbox of MATLAB was employed to make use of neural network for pattern recognition. The networks were built by three layers. The epoch and the learning rate were set to 1000 and 0.01, respectively. The weights were initialized arbitrarily. Further, the network was trained by resilient back-propagation algorithm until the error between the desired and the actual outputs below than the threshold value or until the maximum epoch was reach. Once the weights have been determined, the network can be used as a classifier. The network structure was built based on the experiment by testing several networks model and the designed layers were confirmed as the effective network structure for this problem.

4. RESULTS 4.1 Subjective Evaluation of Human States The questionnaire responds are evaluated by POMS method to measure psychological mood index as shown in Figure 3 and 4 for experiment A and B, respectively. The self-report questionnaire reveals that subjects in the session of pre-experiment and post-experiment. POMS is a psychological test designed to measure a person's affective states. These include tension-anxiety (T-A), depression (D), anger-hostility (A-H), vigor (V), fatigue (F) and confusion (C). Unlike personality traits, profile mood states are thought to be transitory and specific to a given situation, although moods can also be measured for recent prolonged periods such as the past several months. As shown in Figure 3 and Figure 4, there are interesting factors, i.e., vigour (V) and fatigue (F) which showed different scoring between experiment A and B. The “vigour” states represent the condition of your enthusiasm at the time of pre- and past- experiment conducted, while “fatigue” states Thus, this study just evaluated relation between psychological mood index, i.e., vigour and fatigue towards the physiological information.

Figure 3. POMS Results of Experiment A

87

ISBN: 978-972-8924-82-9 © 2009 IADIS

Figure 4. POMS Results of Experiment B

4.2 Performance Evaluation of the Proposed System for Detecting Human States It is believed that physiological information generated by human body have strong relation with psychological condition such as fatigue, vigor and others, which might associate with sensitivity in the kansei engineering fields. However, monitoring human states times by times are difficult to be obtained, even performing by human. Thus, this study attempt to determine human states condition based on the psychological and physiological information by employing several information processing techniques. First, the collected data were subject to analyze using statistical approach. The data were extracted from segment of 3 min data for all signals. Ten statistical features were calculated for each segment of EEG signal and temperature, i.e., the mean, the standard deviation, the slope mean, the maximum and the minimum. To calculate heart rate variability (HRV) features, it used the instantaneous heart rate time series derived from ECG. The ratio of the low-frequency (LF) band (0.04 – 0.08 Hz) and the high-frequency (HF) band (0.15 – 0.5 Hz) were calculated to produce new data.

Figure 5. The performance of neural networks

The LF/HF is used as an index of sympathetic to parasympathetic balance of heart rate fluctuation. The mean, the standard deviation, the slope mean, the maximum and the minimum of the ratio LF/HF were then used as features. HF is driven by respiration and appears to derive mainly from the parasympathetic nervous system. The mean, the standard deviation, the slope mean, the maximum and the minimum of HF were used

88


as final features in this experiment. There 20 features were used to create a single vector representing each of the segments used in the recognition analysis. Totally, 324 segments were extracted from experiment A and B, i.e., 216 and 108 from task and rest periods respectively. The resulting 324 feature vectors were then analyzed by principal component analysis (PCA). Second, the obtained features vectors generated by PCA were reconstructed, so the extracted data belongs to each subject. Totally there were 18 subjects. Each feature vector consisted of 12 features and 18 time series data. The extracted data was then set as training data for HMM network. The parameters of the HMM network were estimated by the given training data. The output of HMM was then used as input training for neural network. The output of training data was consisting of psychological mood differences, i.e., vigour and fatigue, between pre-experiment and postexperiment. Then the same input training was employed to evaluate trained network. The result of trained network was shown in Figure 5. It showed that the correlation coefficient (R) between the output 1 and target 1 (vigour state) showed moderate result, R1=0.6257 while the correlation coefficient R2 (fatigue state) show 0.4480, it means that there are less variation between the output 2 and the target 2. The mean absolute error of classification between target and output are 13.7% and 21.3% for vigour and fatigue, respectively, as shown in Table 1. Evaluating human states by the proposed method showed that the experiment B has lower states i.e., “vigour” and “fatigue” than the experiment A. Evaluating human feeling towards the given task using physiological and psychological information showed promising results in this study. So, there are some possibilities to build systems which have capability predicting human states condition. However, further investigation has to be made to find more clear correlation between that information. Table 1. The results Vigor

Fatigue

Correlation coeficient

0.6257

0.4480

M ean absolute error

13.70%

21.30%

5. CONCLUSIONS This study has proposed a method to estimating human states by understanding physiological and psychological information using information processing techniques. The results showed that the proposed systems are able to estimate “Vigor State” in moderate level, and to estimate “Fatigue State” in medium level. This result is promising for further study relating to understanding human states condition more objectively. However, finding relation between physiological and psychological information is difficult to be achieved in high degree using current method. It might that, first, each person have different capability when handling mental workload. Second, the proposed model is not sufficient enough for evaluating human states condition. Third, the subjective evaluation might not represent the real condition of that person. This research is still preliminary study about relation between accumulations of human states and psychological mood index based on the given task. In future, the systems which are able to explain relation between physiological measurement (objective) and psychological mood index or human feeling, (subjective) should be developed. This information, in the future, could then be used automatically by the adaptive systems in various ways to help the person better cope with stress, fatigue or even for detecting diseases such as heart attack. The example of this might include adaptive systems which are able to analyze the stress level of a person, and give an alert or a suggestion to the person to take a break if required or even sending the critical information to hospital related to the person condition.

89

ISBN: 978-972-8924-82-9 © 2009 IADIS

REFERENCES Francisco Castells, Pablo Laguna, Leif Sörnmo, Andreas Bollmann, and José Millet Roig (2007), “Principal Component Analysis in ECG Signal Processing,” EURASIP Journal on Advances in Signal Processing, vol. 2007, Article ID 74580, 21 pages, 2007. doi:10.1155/2007/74580 Gabriel, P. and Liimatainen, M.R., (2000) ’Mental Health in the Workplace’, International Labor Organization. Healey, J.A. and Picard, R.W. (2005) ‘Detecting Stress During Real-World Driving Task Using Physiological Sensors’, IEEE Trans. on Intelligent Transportation Systems, Vol. 6, No. 2, pp. 156-185. Health, Labor and Welfare Ministry of Japan (2004) ‘Outline of the Result of Actual Condition Survey Concerning Technical Innovation and Labor’ [in Japanese], Obtained through the Internet: http://www.mhlw.go.jp, [accessed 8/2008]. Mizuno, M., Yamada, Y., Mizuno, Y., Matsuda, F., Koizumi, T. and Sakai, K. (2007) ‘An Empirical Study on Work Stress and Health Condition of Japanese Nurses’, Journal on Health and Sports Science Juntendo University, Vol. 11, pp. 58-63. Rabiner L.R. (1989) ‘A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition,’ Proceeding of the IEEE, Vol. 77, No. 2, pp. 257-286. R.O. Duda, P.E. Hart, D.G Stork, “Pattern Classification,” 2nd edition, Wiley-Interscience, 2000. Zhang, C., Zheng, C., Yu, X. and Ouyang, Y. (2008) ‘Estimating VDT Mental Fatigue Using Multichannel Linear Descriptors and KPCA-HMM,’ EURASIP Journal on Advances in Signal Processing, Vol. 8, No. 2.

90


AN APPLICATION OF ROUGH SET ANALYSIS TO A PSYCHO-PHYSIOLOGICAL STRESS RESEARCH Shusaku Nomura Top Runner Incubation Center for Academia-Industry Fusion, Nagaoka University of Technology 1603-1 Kamitomioka, Nagaoka, Niigata 940-2188, Japan

Santoso Handri Top Runner Incubation Center for Academia-Industry Fusion, Nagaoka University of Technology 1603-1 Kamitomioka, Nagaoka, Niigata 940-2188, Japan & Multimedia Nusantara University Tangerang, Indonesia

Yasuo Kudo Department of Computer Science and Systems Engineering, Muroran Institute of Technology 27-1 Mizumoto, Muroran 050-8585, Japan

ABSTRACT This study aims at an application of rough set theory to illustrate the relationship between human psychological and physiological states. Recent behavioral medicine studies have revealed that various human secretory substances change according to mental states. These substances, the hormones and immune substances, show temporal increase against mental stress. Thus, it is frequently introduced as biomarkers of mental stress. The relationship between these biomarkers and human chronic stresses or daily mental states was also suggested in the previous studies. However the results of these studies were sometimes inconsistent with each other. Some technical reasons were indicated for this discrepancy. Among that, we focused on the analysis technique investigating the relationship between human psychological state, i.e., scores of a psychological scale, and physiological state, i.e., level of the secretory biomarkers. In this paper, we introduced Rough Set analysis method in place of a conventional linear correlation analysis method. In the experiment, the salivary secretory immunoglobulin A (IgA), which is a major stress biomarker, of 20 male students was assessed as a physiological biomarker. Also, 65 items of psychological mood scale, “profile of mood states” (POMS), was assessed as a psychological index. The result showed that some items strongly related with the IgA level, while few significant linear correlation was obtained among them. KEYWORDS Rough set, POMS, IgA, Stress.

1. INTRODUCTION In this study, we introduced Rough Set theory (Pawlak, 1982, 1991) as a new methodological approach to extract the embedded relationships between human psychological and physiological states. For over a century, behavioral medicine studies have been made to investigate the relationship between human psychological and physiological states. Consequently, it has revealed that human secretory substances sensitively change accompanying with the change in his/her mental states. For example, human secretory immunoglobulin (IgA) shows transient increase against a short-term experimental stressor like arithmetic task (Valdimarsdottir, 1997; Ader, 2001; and Bosh, 2002). Recently, this bio-behavioral medicine studies are drastically developing according to the improvement of biochemical analysis techniques such as radioimmunoassay (RIA) and enzyme-linked immunosorbent assay (ELISA). Nowadays, at least a dozen of human secretory substances, which are hormones, immune substances, and digesting enzymes, are considered as candidates of the biomarker of human psychological sates (Wakida, 2004). Because most of

91

ISBN: 978-972-8924-82-9 © 2009 IADIS

these biomarkers showed increase against short-term stressors, it is expected to be an objective and practical stress-marker. On the other hand, in contrast to the studies on the short-term laboratory stressors, studies investigating the effects of long-term stressors or rather daily stressful experiences on biomarkers have frequently shown inconsistent results. Some reasons were indicated for this discrepancy, such as small number of studies, the variation of biomarker determination techniques, subjects' control, and specimen sampling timing. Above all, we focused on the fact that in all these studies the only method introduced to evaluate the degree of relationship between somatic and mental state was the correlation analysis, by which a sort of linear-relationship between the score of psychological questionnaire and the level of secretory substance was quantified. The correlation analysis is though a one-to-one factorial evaluation method based on linearity between the target factor and another. In other words one can easily imagine that the correlation analysis would suggest no result if a target factor such as the change in secretory substance were mediated by several psychological factors, or if the relationship were in the form of non-linearity. Therefore it can be worth introducing independent analytical methodology. In this study we introduced rough set theory as an attempt to extract the relationship between the score of psychological questionnaire and the level of a physiological biomarker. In the rough set theory, the factors and elements of data for analyzing have no statistical restriction such as the number of elements, linearity, and independency of factors. Also, relationship among multiple factors can be analyzed simultaneously. In this study we introduced the "Profile of Mood States" (POMS) (MacNair, 1992) as for a psychological questionnaire and the concentration of salivary immunoglobulin A (IgA) as for a biomarker, and analyzed the relation among them by the rough set theory. In the next we briefly review the past IgA studies.

2. SALIVARY IMMUNOGLOBULIN A (IGA) AS A STRESS BIOMARKER Immunoglobulin A (IgA) is one of the most important substances in human immune system (Tsujita, 1999). It works non-specifically and therefore plays a very important role for our health, e.g., preventing bacteria from forming colonies, neutralizing toxins and enzymes produced by bacteria, and inhibiting pathogenic viruses to penetrate into the epithelial cell. This is the reason why salivary IgA called as the "first line of defense" against the influenza or other respiratory tract infection (URTI) illnesses. In fact, clinical studies have suggested the negative correlation between the level of salivary IgA and the incidence of an acute URTI (Jemmott III, 1989). It was also suggested that the relevance between the level of salivary IgA and caries or periodontitis (Gregory, 1992). On the other hand, by the 70's behavioral medicine studies it has been found that salivary IgA changes its level accompanying with various types of psychological factors (Bosh, 2002), such as desirable or undesirable daily events (Stone, 1994), daily hassles (Martin, 1988), negative or positive moods (Martin, 1993), academic stresses such as examination (Jemmott III, 1983) and presentation (Evans, 1994), a shortterm stressful cognitive task (Jemmott III, 1989), and relaxation (Green, 1987; Knight, 2001). A review article has concluded that there are distinguishable two types of stress effects on IgA: 1) increasing IgA secretion immediately after a short-term stress, termed "immediate stress effect", and 2) decreasing IgA secretion several days after or during a long-term stress, termed "delayed stress effect" (Tsujita, 1999). However, even though the immediate stress effect has been successfully observed almost all studies targeting on a variety of short-term stressors, the delayed stress effect has not yet been directly observed. A review paper on IgA studies pointed out that studies targeting on a long-term stress had methodological defects such as less control of subjects' physical conditions like as sleep and diet, using inappropriate saliva sampling methods, and introducing non-standardized psychological questionnaires (Bosh, 2002). So far as we know, no clear relationship between IgA and any long-term stress or daily psychological state has ever shown before. However as mentioned above, these studies frequently investigated the relationship solely by the correlation analysis, which was based on the linearity of one factor to another. Because scores of questionnaire are subjective, there must be a space for contaminating some sort of uncertainty. We then introduced the rough set theory which was a non-parametric analysis method and was thought to be suitable for classifying the data sets including such uncertainty.

92


3. METHOD Twenty healthy male students aged from 20 to 26 participated in this study. They were well informed about the aim and contents of this study before the experiment, and confirmed their participation by subscribing to the agreement. Subjects’ saliva was taken by small cotton placed under their tongue for 3 min, and kept in a freezer at 20 Celsius by the day on biochemical assay. The concentration of IgA was determined by the enzyme-linked immunosorbent assay (ELISA). With regard to the psychological test, subjects were required to fill up POMS (MacNair, 1992; Yokoyama, 1993). POMS is one of the major questionnaires frequently introduced various psychological studies. It consists of 65 items asking about subjects' mood with five point scale: not at all, a little, moderately, quite a lot, and extremely. These items are designed to classify into the six identified mood factors: tension-anxiety (T-A), depression-dejection (D), anger-hostility (A-H), fatigue-inertia (F), vigoractivity (V), and confusion-bewilderment (C). The score of each mood factors is obtained by adding the corresponding items. Also, there are seven items which are not classifiable into the mood factors (dummy items). We analyzed the relationship between IgA concentration and the score of POMS of the subjects by the rough set analysis. When we conducted the rough set analysis, only IgA was assigned as the decision attribute as the simplest attempt, and scores of POMS corresponding to each mood factor were assigned as condition attributes. The idea of rough set theory including our procedure of analysis is briefly described in Appendix A.

4. RESULT As a result of biochemical analysis, the average IgA concentrations was 123.5 [μg/dL] (SD = 63.5). Note that the IgA of a subject was lost because of the failure of biochemical determination procedure. The rest data obtained from 19 subjects were introduced to the correlation and rough set analysis. Table 1 shows the correlation coefficients between IgA concentration and the six identified mood factors of POMS. As the table shows, there is no clear correlation between IgA and mood factors of POMS except for factor C (confusionbewilderment). Taking into account the assumption that the salivary IgA can be a useful stress biomarker, it is incomprehensible that the factor C solely had a negative correlation with IgA, whereas other factors had no significant correlation. As a whole, it was suggested that there was no linear relation between POMS and IgA just as the past review paper concluded. With regard to the rough set analysis, we categorized IgA at regular intervals for applying the rough set (non-parametric) analysis as Table 2 shows. Table 3 shows the result of rough set analysis, which are the degrees of contribution (DoC) (see Appendix A) and the correlation coefficients between each items of POMS and IgA. It also shows the average DoC of the six mood factors and dummy items. The correlation coefficients are shown only if they are statistically significant (p< .05). There were only two items showing statistically significant correlation, q5 and q32 in factor C. Note that such a correlation was a result of simple one-to-one correlation between each item and IgA. Therefore, it is not necessary to think that the items with higher correlation had closer relationship with IgA than the other items. As the result of Table 1 taken together, it is suggested that almost all the items and factors of POMS might not have relevant linear relation with IgA. On the other hand, there were several items with higher DoC ( ≥ 0.6) such as the item q53 in T-A, q51 in D, q41 and q52 in A-H, q4 and q19 in V, q9 and q51 in F, and q5 in C. Moreover, the dummy items q1, q30, and q31 also had higher DoC. Table 1. Correlation coefficients between IgA and each factor of POMS F C T-A D A-H V IgA -0.22 -0.32 0.00 0.04 -0.41 -0.47* * p ma AND b > mb • Group 2 – Potential promoters: a > ma AND b ≤ mb • Group 3 – Hidden opponents: a ≤ ma AND b > mb • Group 4 - Opponents: a ≤ ma AND b ≤ mb The chart (Figure 1) displays the assignment of the individuals of each school to the four resistance groups. It is interesting that School 1, which is located in a Faculty related to health science, has the highest amount of opponents or hidden opponents – in total 61%. Compared to the other schools, in a scienceoriented faculty, the information and communication measures seemed to be quite good. A reason for the higher degree of resistance might be that this school is situated in the health sciences and is the one with the least focus on software systems.

191

n=10 n=8 n=8

n=12 n=2 n=0

n=6 n=5 n=6

n=8 n=6

50.00% 40.00% 30.00% 20.00% 10.00% 0.00%

n=8

% of school

ISBN: 978-972-8924-82-9 © 2009 IADIS

S chool 1 S chool 2 S chool 3

P romoters P otential Hidden O pponents P romoters O pponents R es is tanc e G roups Figure 1. Resistance groups relating to Schools

6. CONCLUSION This study suggests that a quantitative analytic method used in association with Krüger’s Iceberg Model has the potential to transform the model into a useful tool for measuring and quantifying attitudes and behaviours. We have called the method Iceberg Model Analysis (IMA). Using IMA in this study revealed a number of factors that affect attitudes and behaviours in relation to resistance to change. These factors include: perception of how work characteristics are investigated and how they inform system development, training on the system, as well as the information about reasons that induce the need for change and how the change will influence daily work. Influenced by these factors individuals form their opinions and become promoters or opponents. IMA helps to identify the nature and extent of resistance groups. Armed with better knowledge of resistance groups the change management team will be able to take more informed decisions about their approach to the management of change. How resistance to change can be overcome is not a part of this study.

ACKNOWLEDGEMENT We acknowledge the support of those that provided information about the introduction of the workload model and thank all participants who gave up their time to participate in the study reported here.

REFERENCES Anderson, D., Ackerman-Anderson, L., & Anderson, D. (2001). The Change Leader's Roadmap: How to navigate your organization's transformation. San Francisco: Jossey Bass. Brown, D. (2008). Work Load Modelling - A time line. Edinburgh: University's Project Management. Joint Costing and Pricing Steering Group (2005). Transparent approach to costing. Higher Education Founding Council for England. Krüger, W. (2004). Implementation: The core task of change management. In B. de Wit, & R. Meyer, Strategy - Process, Content, Context. London. Mittelhammer, R. C. (1996). Mathematical Statistics for Economics and Business. New York: Springer.

192


INDIVIDUAL “ICT LANDSCAPES” AND QUALITY OF LIFE Pernilla Gripenberg Hanken School of Economics Arkadiankatu 22, P.O. Box 479, FI-00101 Helsinki, Finland

ABSTRACT Everyday life is becoming increasingly ‘technologised’ and ‘virtualised’ through the growing use of various information and communication technologies (ICTs) across all spheres of life. ICT devices are becoming smaller and more mobile, and also converging, including more different types of functions. Simultaneously, working adults who are evidently more stressed for time than previous generations are also struggling to cope with a growing number of ICTs that do not necessarily ‘talk’ smoothly with each other, leaving people ever more dependent on technological support and the good will of technology interested friends. On the other hand, the general assumption is that ICTs are making our lives better in terms of saving time, increasing mobility, increasing the amount of information we can access, decreasing the need for travel etc. Previous research has clearly shown that there are many paradoxical effects of ICT usage and the question of whether or not the use of ICTs are increasing our quality of life, well-being, life control and personal autonomy is not possible to answer straightforwardly. In this short paper I offer a résumé of key issues from previous and ongoing research and propose some avenues for future research on individual ICT landscapes and quality of life. KEYWORDS ICT use, ICT landscapes, quality of life, multi-method, sociomateriality, social informatics

1. INTRODUCTION Today, ICTs are not only used in work contexts, but increasingly ICTs are becoming mobile and being used across a variety of everyday contexts, such as at home, during travel, when abroad, in a net café or hot spot. They are not only used for work purposes, but as much for private and leisure activities, such as staying in touch with friends, playing games, filming and photographing, editing and displaying photos and videos etc. Increasingly, people have portable PCs with wireless internet access (WLAN), Black Berries and other portable devises that are used in combination or separately from the mobile phone. The mobile phone is not even only a mobile phone any more: it is a camera (both still and video), a calendar, a radio, a ‘communicator’ with word processing and e-mail features, a game, it is even used as a flash light, an alarm clock, etc. Simultaneously, new ICT products with integrated features are constantly being developed and entering the market in a steady stream. ICT that for the average person is increasingly difficult to integrate with the growing amount of older technology, is making life with ICT ever more complicated. Based on the author’s extensive doctoral research in the area of ICTs in everyday life contexts (Gripenberg, 2002, 2004, 2005 and Gripenberg et al. 2004) and inspired by the notion of archaeological layers of ICTs (Suchman et al. 1999), this paper seeks to suggest new avenues for researching and understanding the ICT landscapes that surround individual ICT users in their various everyday life contexts. The suggested research agenda is to first, identify and describe the layers of ICTs, not only in work or home contexts, but in the various places where people are and through which they move during a certain period of time. Second, it is to explore the effects of these ICT landscapes on individuals’ perceived quality of life, life control and well-being.

193

ISBN: 978-972-8924-82-9 © 2009 IADIS

2. PREVIOUS RESEARCH AND SUGGESTED RESEARCH FRAMEWORK A number of ethnographic studies of technology use and in particular ICT use in the work context have been conducted over the past 30 years (Suchman 1987, Zuboff 1988, Orr, 1996). In an article summarizing 20 years of the authors’ ethnographic exploration of the relations between everyday practices and technology design and use in various contexts of work, Suchman et al. (1999) conclude: “rather than the primacy of one technology over all others, powerful technical systems comprise artful integration…it is through the infrastructural details that the texture of our technology-infused society is woven.” In their study of an airport operations room they discovered that rather than forming a homogeneous and integrated network of ICTs, the technologies in use presented a kind of archaeological layering of heterogeneous artifacts acquired in bits and pieces over time. Integrating them then seemed more like a matter of string and bailing wire than of design. At present a small but emergent body of socio-cultural ICT research where empirical studies have integrated the use (and included also non-use) of several ICTs or different contexts into the same study can be identified in the works by Frissen (2000), Green (2002), Selwyn (2003), Leung and Lee (2005), Jokinen (2005), Peters and Allouch (2005), Anderson (2007), Leung et al. (2008), Schwanen and Kwan (2008) and Wajcman et al. (2008). However, the majority of studies focusing on ICT use have traditionally taken a single technology and/or single context perspective (for review on IT in the home context see National Science Foundation, 2001 and for reviews on IT in work organization see e.g. Orlikowski and Baroudi, 1991, Liker et al. 1999. Orlikowski and Scott, 2008). In the Finnish context, many are focusing directly on how to increase ICT use and uptake e.g. in consumer and marketing research much focus has been put on how to develop mobile services and e-commerce (e.g. Carlsson and Walden 2002, Pihlström, 2008, Sintonen, 2008) whereas it is less common to ask what the usage really means for the users. Single technology or single context studies have generated much new, but often contradictory findings about the outcomes of ICT use and thus have limitations in further developing our knowledge in this area. Another major issue in this research area is the limitations of the traditional and often technology focused research approaches (for further discussion see e.g. Robey 1997, Kling 2000, Woolgar 2002). In reply to the call for new approaches the interdisciplinary area of social informatics research (Kling 2000, Sawyer and Escenfelder 2002) is growing in importance. Social informatics research provides us with important means for increasing our awareness and for broadly discussing ICTs’ varied influences in society. New approaches from this area include perspectives on technology from the interpretivist/social constructivist tradition that can be used as sensitizing devices to guide more grounded analysis of the ICT-user relationship in the context were the use occurs. New perspectives in this area include recent developments in ideas as sociomateriality (Orlikowski, 2007), actor-networks (originally developed by Callon, (1986) and Latour, (1987)) and apparatgeist (Katz and Aakhus, 2002). These perspective have some central ideas in common, maintaining a view on technology were interpretive flexibility is possible, yet not infinite. They also build on the relational ontology that does not separate people from things but regards them as inherently inseparable (Orlikowski and Scott, 2008). Building on these recent developments and previous research in this research area a suggested framework for future research is summarized in Fig. 1.

194


1. Societal prerequisites for ICT use including • Penetration rate of ICT (private and organizational) • Longevity of ICT • Demographics

Theoretical perspectives for studying human-ICT relationships including • sociomateriality • actor-network theory • apparatgeist

2. Individual use of different ICTs in various contexts, including: • work, • home and • mobile use of ICTs

3. Perceived quality of life, wellbeing and life control in relation to ICT use: Including: • material • social, • physical and • emotional wellbeing • development and activity

Figure 1. Suggested research framework for future research

3. AVENUES FOR FUTURE RESEARCH Based on previous research and the shortcomings presented here some avenues for future research are proposed. First, further attempts should be made to combine the study of different ICTs that have previously mostly been studied in separation, like IT or mobile phones. Second, attempts to study the use of ICTs in a combination of various life settings, not separating home, work and leisure contexts from each other should be made. Third, as ICTs are often ascribed life enhancing qualities when marketed (that for various reasons are not self-evident in the practical use of ICTs), the social and behavioral effects of the use of ICTs in terms of quality of life (for reviews, see e.g. Anderson 2004, Moons et al. 2006), well-being and life-control should also be addressed in future research. Further, a future research agenda should combine areas of information systems research with organization, and communications research as well as with social research on well-being and quality of life. As such new findings would inform the longer-term developments of ICT use in authentic everyday life contexts. Such studies would generate new knowledge on everyday life in the information society. Common to the results in the social informatics field is that they also often yield normative recommendations to professionals who design, implement, use or make policy about ICTs (Sawyer and Eschenfelder, 2002). By providing detailed empirical descriptions of the varied outcomes that occur as people use different ICTs in combinations across a variety of context – the “ICT landscapes” of people’s everyday life, would also provide some suggestions for design and policy making as well as implementation and use. Methodologically, suggestions for future research could include multi-method approaches triangulating various data gathering methods to tap into routinised and taken for granted patterns of behavior in relation to ICT use across various settings. Taken for granted patterns of behavior might otherwise be difficult for subjects to express and therefore difficult to detect using single method approaches (Frissen 2000, Punie 1997). In her study of ICT use of dual career families in the Netherlands, Frissen (2000) used some interesting methods for contextualizing as much as possible the ICT use of the studied family members. Frissen combined in-depth interviewing and observation, with semi-structured questionnaires, ICT-use diaries, mental mappings of the house (to visualize the physical space in which ICTs were placed), and network diagrams (to make visible the social networks the households were embedded in). Also Jokinen (2005) used a mental mapping method in her study on the everyday lives of Finnish adults. Future research in

195

ISBN: 978-972-8924-82-9 © 2009 IADIS

this area could build on these methods, adding for example the use of video to map the ICT landscapes of individual ICT users.

ACKNOWLEDGEMENTS I am most grateful for the valuable comments on this paper by prof. Linda McKie at Glasgow Caledonia University and for comments on this research by my colleagues Dr. Liisa Husu and prof. Jeff Hearn at Hanken School of Economics.

REFERENCES Anderson, B. (2004): Information society technologies and quality of life – a literature review. Chimera Working Paper 2004-09. Ipswich: University of Essex. Anderson, B. (2007): Using time use data to analyse macro and microsocial change in an e-society: Full research report. ESRC End of Award Report, RES-341-24-0004. Swindon: ESRC Callon, M. (1986). Some elements of a sociology of translations: Domestication of thescallops and the fishermen in St Brieuc Bay. In J. Law (Ed.), Power, action, and belief: A new sociology of knowledge. London: Routledge. Carlsson, C. and P. Walden. Further Quests for Value-Added Products & Services in Mobile Commerce. In Proceedings of the ECIS 2002 - The European Conference on Information Systems , 2002. Easterby-Smith, M., Thorpe, R. and Jackson, P.R. (2008): Management Research. Third Edition. London: Sage Publications Ltd. Frissen, V. (2000): ICTs in the rush hour of life. The Information Society, vol 16, p. 65-77. Green, N. (2002): On the move: technology, mobility, and the mediation of social time space. The Information Society, vol. 18 p. 281-292 Gripenberg, P. (2002): Living with IT: Uses and interpretations of computers in the home and family context. Proceedings of the Xth European Conference on Information Systems, June 6-8.2002, Gdansk, Poland pp. 1261-1272. Gripenberg, P. (2004): Virtualizing the office: micro-level impacts and driving forces of increased ICT use. In Heiskanen, T. and Hearn, J. eds. (2004): Information Society and the work place: Spaces, boundary and agency. London and New York: Routledge. Gripenberg, P. (2005): ICT and the shaping of society: Exploring human-ICT relationships in everyday life. Doctoral thesis at Swedish School of Economics and BA. Helsinki: Edita Prima Ltd. Gripenberg, P. and Skogseid, I. with Botto, F., Silli, A. and Tuunainen, V.K. (2004): Entering the European Information Society: Four rural development projects, The Information Society, Vol. 20, no. 1, pp.3-14. Jokinen, E. (2005). Aikuisten arki. (The everyday life of adults) Helsinki: Gaudeamus Kling, R. (2000): Learning about information technologies and social change: The contribution of social informatics. The Information Society, vol.16. p. 217-232. Latour, B. (1987). Science in action. Boston, MA: Harvard University Press. Lee, P., Leung, L., with Venhwei, L. and Chengyu, X. (2008): The perceived role of ICTs in Quality of Life in Three Chinese Cities. Social Indicators Research, 88(3), September, p. 457-476. Leung, L. and Lee, P.S.N. (2005): Multiple determinants of life quality: the roles of Internet activities, use of new media, social support, and leisure activities. Telematics and Informatics, vol. 22 p. 161-180. Liker, J.K., Haddad, C.J. and Karlin, J. (1999): Perspectives on technology and work organization. Annual Review of Sociology, 25: 575-96. Moons, P., Budts, W. and De Geest, S. (2006): Critique of the conceptualization of quality of life: A review of different conceptual approaches. International Journal of Nursing Studies, Vol. 43, No. 7, p. 891-901. Orr, J. (1996): Talking about machines: An ethnography of a modern job. Ithaca, New York: ILR Press. Orlikowski, W.J. (2007). Sociomaterial practices: Exploring technology at work, Organization Studies, 28, 1435–1448. Orlikowski, W. and Scott, S. (2008): Sociomateriality: Challenging the separation of technology, work and organization. The Academy of Management Annals, Vol. 2, No. 1, 2008, 433–474. Peters, O. and Allouch, S. (2005): Always connected: A longitudinal field study of mobile communication. Telematics and Informatics, vol. 22, p. 239-256.

196


Punie, Y. (1997): Rejections of ICT in Flemish households. The why-not question. In EMTEL working paper no. 3. Media and Information technology: Regulating markets & everyday life eds. R. Silverstone and M. Hartman, pp. 4672. Brighton: University of Sussex. Pihlström, M. (2008): Perceived value of mobile service use and its consequences. Doctoral thesis at Swedish School of Economics and BA. Helsinki: Edita Prima Ltd. Robey, D. (1997): The paradoxes of transformation, in eds. Sayer, C. and Yetton, P.W.: Steps into the future: Fresh thinking on the management of IT-Based Organizational transformation. San Francisco: Jossey-Bass Publishers. Sawyer, S. and Eschenfelder, K. (2002): Social Informatics: perspectives, examples and trends. Annual Review of Information Science and Technology. Schwanen, T. and Kwan, M-P. (2008): The Internet, mobile phone and space-time constraints. Geoforum, vol. 39, p. 1362-1377. Selwyn (2003): Apart from technology: understanding people’s non-use of information and communication technologies in everyday life. Technology in Society, vol. 25, p. 99-116. Sintonen S. (2008): Older consumers adopting information and communication technology: Evaluating opportunities for health care applications, Acta Universitatis Lappeenrantaensis 326 Suchman, L. (1987): Plans and situated actions: The problem of human-machine communication. New York: Cambridge University Press. Suchman, L., Orr, J. and Trigg, R. (1999): Reconstructing technologies as social practice. American Behavioral Scientist. vol. 43, No. 3, November/December p. 392-408. Wajcman, J., Bittman, M. and Brown, J. (2008): Families without borders: Mobile phones, connectedness and workhome divisions. Sociology, Vol. 42 No. 4. 635-652. Woolgar, S. (ed.) (2002): Virtual Society? Technology, Hyperbole, Reality. Oxford: University Press. Zuboff, S. (1988): In the age of the smart machine: the future of work and power. New York: Basic Books.

197

ISBN: 978-972-8924-82-9 © 2009 IADIS

THE ROLE OF INFORMATION SYSTEMS IN INFORMATION MANAGEMENT: EXPLORING THE CHALLENGES IN INFORMATION SHARING Seppo Sirkemaa Turku School of Economics, Pori Unit P.O.Box 170 Pori – Finland

Karen Anderson, Karl W. Sandberg, Olof Wahlberg, Erik Borglund Mid Sweden University S851-70 Sundsvall - Sweden

ABSTRACT Information systems may be used for several purposes in an organization. They are often the basis of operations and activities. For example, information systems and technologies can be an integrated part of the production process. Information systems have also an important role in management and development of the organization. Here information, its management and sharing throughout the company have a decisive role. It may still be argued that in many cases information systems are not used in the best possible way. For example, a lot of information remains personal property even though it might be valuable when shared with colleagues. In this paper we look at the role and use of information systems in information sharing. The paper explores the challenges in sharing of information and knowledge. Technology can be used here, especially if the information can be expressed in written format. When information is more tacit, complex and based on expertise also personal interaction is important in information sharing. KEYWORDS Information systems, Information management, Knowledge sharing, Development

1. INTRODUCTION Information management is a key issue in organizations. Clearly, information is needed in order to run operations and develop the organization further. In smaller organizations the entrepreneur tends to have a hands-on touch on operations and knows how to run the company. The development of products and services does, however, require information on customer needs, changes in environment and other similar issues. These call for infrastructures and procedures for information management and sharing. Information management becomes even more important when organizational size increases. When there are several people in the company there is a need to know what others are doing. Sharing information is needed between individuals, work teams and offices – all should have access to knowledge and information throughout the organization. Consequently, platforms and mechanisms for managing, storing and sharing information become core issues in the organization.

1.1 About the Paper This paper discusses information management and sharing in organizations. Information management is here understood as managing, storing and sharing information. There are several practices, methods and mechanisms for managing information in organizations. As an example, industry and organizational size are factors explaining the variety of different knowledge sharing methods in companies. Also organizational culture impacts information management practices, therefore otherwise similar organizations may have very

198


different information sharing practices. The challenge is finding the best method in sharing information in each organization. In this research we look at information management infrastructures and practices. The paper is built on information management frameworks and knowledge-sharing mechanisms. They are used in studying different knowledge sharing practices found in organizations. The practical goal of the researchers is to increase the awareness of the potential of information systems and to better take advantage of the systems. The empirical part of the research is currently in progress – detailed results will be available in a few months.

2. INFORMATION MANAGEMENT Information management is a multifaceted issue. It may be seen as management of data – storing and managing bits and bytes. The approach to information management can be strictly technical concentrating on computer hardware and infrastructures. On the other end of the continuum information management deals with sharing information and knowledge. It is closely connected to company management, setting strategy and sharing vision throughout the company. A critical issue – managing information and sharing it can give the company competitive edge and increases the chances of success (Nunes et al. 2006). This leads to the fact that information and knowledge need to be shared in organizations. Depending on the type of information there are different mechanisms for doing this. From a technical perspective there are different hardware, software and infrastructure-related solutions available which are intended for storing and sharing data.

2.1 From information to knowledge Information management can be seen as management of information, and sharing this information with others so that it adds value. This requires often understanding what information is about, contextual factors are important so that information can be used wisely. Here context is the element that separates information from knowledge (Blair 2002). Knowledge-sharing mechanisms are here defined as either formal or informal mechanisms which are used for sharing knowledge embedded in individuals or groups so that it could be used in various processes and activities. However, the information that should be shared may not always be easily shareable. Let us look at these issues in more detail. Firstly, not all data and information is meant to be shared (Olivera 2000). Individuals may generate, store and work with information that is meant for “private use”. On the other hand, in organizations there is also plenty of information that is produced together with others and shared among colleagues. This type of shareability is referred to as aggregation of knowledge (Nonaka 1994, Gabrera and Gabrera 2002). Aggregation of knowledge varies from individual to collective. By definition, individual knowledge and information is known by one person alone and therefore private by nature. Collective knowledge is intended to be shareable and available for several persons, for example to colleagues or team-members. Another viewpoint in knowledge and information sharing is the degree of articulation (Polanyi 1962). In general, some types of knowledge are relatively easy to express in oral or written format. In organizations there is plenty of data which is generated in organizational processes and directly stored in written format, which makes it easy to use, combine and transfer this data from one location to another. As an example, data on production volumes and sales is a result of these activities, and the related data is typically stored to information systems which are used in these functions. For example: How many units of product A have been manufactured in line X on last Tuesday? This kind of data and information is the result of the manufacturing process, straightforward to understand and therefore can be easily shared with others. On the other hand, there is also knowledge which can be more difficult to articulate, often because it deals with issues which are not straightforward. Why have the production volumes of line X been dropping? The example indicates the difference of challenges related to articulation. The first question can be answered by checking the production numbers. The prerequisite is that there is a system for storing, browsing and retrieving manufacturing information. However, the second situation – dropping production volumes can be the result of numerous issues which have their roots inside the company or its environment. For example, production volumes may be lower because there is not enough demand on the customer side. The

199

ISBN: 978-972-8924-82-9 © 2009 IADIS

explanation needs reflection, combining viewpoints and sharing insights. In many cases expertise and understanding factors that affect manufacturing is needed in order to answer the question. This is referred to as tacit knowledge (Nonaka 1994). In knowledge sharing both the information that is to be shared and the characteristics of those who are sharing the information are relevant. In personalization the absorptive capacity is of importance, it refers to the information seekers prior knowledge of the subject area which in turn helps in understanding the information. Another important characteristic is the perceived reliability of the information source. The third aspect is the interaction of those who are sharing information.

3. KNOWLEDGE SHARING MECHANISMS Generally, knowledge sharing is based on two mechanisms: knowledge is shared from person-to-person which is referred to as personalization, or using technical solutions. It is often argued that knowledge sharing mechanisms that are based on personalization would be rather informal and ad hoc. In contrast, technical solutions are assumed to be more formal and incorporate electronic databases in knowledge sharing (Choi and Lee 2003). As such information can be classified into four main types (Nonaka 1994): individual-explicit, individualtacit, collective-explicit and collective-tacit. Information management requires often information to be in a form that is explicit to others, meaning that it has to be articulated and then shared in a understandable form. As a result, information has to be converted from one type to another (for example, individual-tacit knowledge must be converted to a format that can be shared with colleagues). (Nonaka 1994, Crossan et al. 1999). Mechanisms for sharing knowledge are needed. When knowledge and information can be expressed in words it can be shared in written format, transmitted and stored in electronic format which allows retrieval and combination of information. The mechanisms that are related to making unclear, non-verbal information shared among others are not as straightforward. For example, non-verbal information cannot be directly codified to a repository that allows browsing and retrieving of information. Also in this case organizations have to develop platforms for sharing knowledge, mechanisms that facilitate person-to-person knowledge-sharing are needed. This takes place through social networking (Hansen 1999). The advantages of social networking as a knowledge sharing mechanism are related to flexibility, and the possibilities for transmitting and sharing of tacit knowledge. Discussions and sharing different viewpoints can also result in development of new knowledge (Prencipe and Tell 2001). Knowledge sharing through personalization tends to have limited reach, or there are few people who may get access to the knowledge (Ewans and Wurster 1997). Knowledge sharing via personalization requires that the person who seeks knowledge is aware of what information others have, where to find memos, working papers etc. Furthermore, the knowledge seeker has to get into touch with the knowledge provider, and the knowledge provider should be willing to share knowledge with the knowledge seeker. There are also other concerns; seeking information from others in the company may be considered as admitting ignorance on a given issue (Menon and Pfeffer 2003). It is still important to facilitate and empower person-to-person knowledge sharing, especially if the information is tacit. Knowledge codification refers to knowledge-sharing mechanism which aims to capture knowledge into knowledge bases that are accessible to others in the company. The mechanism should capture individual or collective knowledge making it wider property of the organization (Earl 2001). The knowledge-sharing mechanisms that rely on codification are typically based or take advantage of information technology and knowledge management applications in converting knowledge into a format that allows archiving, browsing and retrieving information. Technical infrastructures for sharing codified knowledge need to be purchased, implemented and developed in order to create a platform for sharing this type of knowledge.

200


Table 1. Framework for sharing knowledge Individual

Collective

Tacit - Personalization

Social Networking

Explicit - Codification

Sharing personal memos, working papers etc.

Facilitating person-to-person knowledge sharing Knowledge management applications

The framework for sharing knowledge (Table 1) connects the two main dimensions of information and knowledge. Firstly, there is the level of aggregation - individual or collective - and secondly, the degree of articulation – ranging from tacit to explicit. The degree of articulation on the axis explicit – tacit on one hand, and the level of individualization and collectivism on the other hand become evident in the mechanisms that are used in sharing information. Expressed differently, the information as such is the element which leads us to use different methods in sharing information. Information technology has an increasingly important role in knowledge sharing, but it alone is not enough. Boh (2008) argues that when complexity of information increases more personal interaction is needed in understanding it. This highlights the importance of personal interaction in this context. When information is explicit, and/or collective by nature storing, retrieving and transferring it is relatively easy using information technology (down-right corner in Table 1). Information that is individual and at the same time tacit needs most elaboration before it can be shared (the opposite corner in Table 1). (Crossan et al. 1999, Cabrera and Cabrera 2002).

4. KNOWLEDGE SHARING MECHANISMS In order to better understand information management and the challenges in information sharing in organizations we approached the issue with an empirical study. We gathered a group of case organizations from Sweden and Finland and approached selected case organizations with a qualitative research agenda. The research is called UniPlatform (more information on research in progress can be found from the site http://www.samk.fi/uniplatform ). Preliminary notions indicate that information systems are not very well used – in the sense that there is a lot of potential and features in existing systems in organizations that are not currently used. Most information sharing relies on person-to-person interaction. Sharing documents is typical, for example when discussing projects with customers. Social networking is also an important role. In many cases the full potential of information technology is not in use.

5. CONCLUDING THOUGTHS Information management - storing, retrieving and sharing information - is important in all organizations. The key challenge here is sharing information, making it more widely known throughout the organization. Often people generate knowledge in the process of doing their work, and this knowledge tends to remain “personal” or stored in the memory aids that people create for themselves (Olivera 2000). However, this knowledge and information should be shareable and accessible to others. Knowledge is also integrated, embedded and institutionalized in structures and routines of the organization. It is important to notice that people are in the core of all knowledge sharing mechanisms – not technology even though there were all kinds of technical solutions available. Obviously, information technology is a tool for information management. Technology is particularly suitable for management and sharing of information that can be captured, stored and transferred in written format. Technology “fits” to management of explicit

201

ISBN: 978-972-8924-82-9 © 2009 IADIS

information, but tacit, experimental knowledge cannot be treated similarly. Here social networking and person-to-person interaction are needed. Managing experimental, tacit knowledge and making it more publicly available calls for platforms, structures and practices. Different kinds of knowledge sharing mechanisms are needed - they complement each other in information sharing.

REFERENCES Boh, W. F., 2008. Reuse of knowledge assets from repositories: A mixed methods study. Information & Management, 45, pp. 365-375. Blair, D. C., 2002. Knowledge management: Hype, hope, or help? Journal of the American Society for Information Science and Technology, 53(12), pp. 1019-1028. Cabrera, A., and Cabrera, E. F., 2002. Knowledge-sharing dilemmas. Organization Studies, 23(5), pp. 687–710. Choi, B., and Lee, H., 2003. An empirical investigation of KM styles and their effect on corporate performance. Information & Management, 40(5), pp. 403–417. Crossan, M. M., Lane, H. W., and White, R. E., 1999. An organizational learning framework: From intuition to institution. Academy of Management Review, 24(3), pp. 522–537. Earl, M., 2001. Knowledge management strategies: toward a taxonomy. Journal of Management Information Systems, 18(1), pp. 215–233. Evans, P., and Wurster, T., 1997. Strategy and the new economics of information. Harvard Business Review, 75(5), pp. 71–82. Hansen, M. T., 1999. The search-transfer problem: the role of weak ties in sharing knowledge across organization subunits. Administrative Science Quarterly, 44(1), pp. 82–111. Menon, T., and Pfeffer, J., 2003. Valuing internal vs. external knowledge: explaining the preference for outsiders. Management Science, 49(4), pp. 497–513. Nonaka, I., 1994. A dynamic theory of organizational knowledge creation. Organization Science, 5(1), pp. 14–37. Nunes, M. B., Annansingh, F., Eaglestone, B., and Wakefield, R., 2006. Knowledge management issues in knowledgeintensive smes. Journal of Documentation, 62(1), pp. 101-119. Olivera, F., 2000. Memory systems in organizations: an empirical investigation of mechanisms for knowledge collection, storage and access. Journal of Management Studies, 37(6), pp. 811–832. Polanyi, M., 1962. Personal knowledge: Toward a post-critical philosophy. New York: Harper Torchbooks. Prencipe, A., and Tell, F., 2001. Inter-project learning: processes and outcomes of knowledge codification in projectbased firms. Research Policy, 30(9), pp. 1373–1394.

202


SOME GUIDELINES FOR THE DESIGN OF INFORMATION SYSTEMS FOR COMPLEX DECISION MAKING: PRELIMINARY RESULTS Pascal Vidal CERAM - Rue Dostoïevski – BP 085, 06902 SOPHIA ANTIPOLIS Cedex, France.

François Lacroux Université de Toulon -IUT de Toulon, BP 132, 83957 LA GARDE Cedex, France.

Darek M. Haftor University of Växjö, Sweden

ABSTRACT The text presents some preliminary results from an inquiry into information systems for the support of decision-making, particularly in complex situations. A constructivist notion of complexity is assumed followed by H.A. Simon’s canonical model of decision-making, all as a foundation for the elaboration of a set of guidelines for the design of decision information systems. Two key theses are evolving in this inquiry: (i) a holistic understating of decision making processes, including its various cognitive functions and loops, (ii) the representation and intermediation of complex situations to be prioritised before the optimisation and resolution of decisions. KEYWORDS Complexity, decision-making, cognitive functions, representation, intermediation

1. INTRODUCTION This text presents some preliminary findings form an ongoing research into decision-making and its Information Systems (IS). More specifically, a set of preliminary guidelines are proposed for the design of IS with the aim to support decision-making processes in complex human affairs situations. Our position is that in such situations, most of IS:s conceived today have an unexplored potential of contributing to the constitution of decision-making processes with capabilities that cannot be realised otherwise, without decision-oriented IS.

1.1 The Need of Information Systems for Complex Decision-Making In a discussion about the future of Information Systems, the Noble Prize Laureate H.A. Simon (1997) emphasised two key points for further development of IS. One was the general misunderstanding of the very purpose of an IS. While most of effort is put to conceive IS that focus on the information itself, Simon says that we forget that intrinsic value of any information is much less important than its relevancy for decisionmaking. Therefore, he maintains, most of IS:s are information-oriented in their functionality rather than decision-oriented, and this is the reason for why IS did not keep its early promises and have only a minor impact on the decision-making processes. The second key point highlighted by Simon, is the still prevailing syntactic notion of information. In the Shannonian shadow, the focus is on the quantity and the physical signal as such, as well as its channels rather than on quality of information, its semantic meaning, as well as its representations and interpretations (e.g. Fulk & DeSanctis, 1995). As a consequence of this misconception, Simon argues that another wrong assumption influence the design of today’s IS; it is the assumption that the (lack of) information is the

203

ISBN: 978-972-8924-82-9 © 2009 IADIS

bottleneck, while Simon (1997, p.5) observes that “information is not the scarce resource; human time and attention is the scarce resource”. Putting Simon’s two observations together, suggests that we need to focus our attention on the decisionmaking process as such and also on the decision-making actors, rather than on the information processing capabilities and on the information volume. If we turn our attention shortly to the understanding of the decision-making, it seems that a variety of quarters suggest a similar trend. The traditional notion of one complicated yet fully comprehensive human world and therefore of the possibility of optimal decision-making outcomes within it, as sprung out of Enlightenment, is in the middle of its replacement. The emerging understanding seems to be that of the paradigm of complexity. In short, it suggests a plurality of unpredictable and emerging human-worlds, that at best may be locally intelligible, and therefore the possibility of satisfactory decision-making only, at best. Relating the two areas together – the need for re-conceptualisation of the role of an IS, and the emerging notion of the complex world – a seemingly paradox emerges: how can we even assume the possibility of developing tomorrows IS in a decision-oriented mode, when the very decision-making situation denies the possibility of optimal decisions? Our position here is that this seemingly paradoxical situation is precisely a key motive for the need to research into the decision-oriented IS, and that the IS present an unexplored potential, even greater in the world of complexity. The here investigated research domain may thus be summarised in the following general question: what conditions should guide the design of decision-oriented IS in order to support the decision-making process in complex human affairs? For the practitioner of IS design, we offer a set of design guidelines that moves the focus toward the decision-oriented IS, while for the researchers we offer a set of hypotheses for further elaboration. The proposals presented here are still in their hypothetical phases, and thus in need of empirical experience and testing.

2. RECEIVED THEORIES This part presents some key findings from the Complexity Sciences and the Decision Sciences as a foundation for the further elaboration of the decision information systems.

2.1 Emergent Behaviour: Insights from Theories of Complexity There is a variety of notions of complexity and of complex systems (e.g. Weaver, 1948; Kauffman, 1993; Holland, 1998) however this elaboration relies on Le Moigne’s (1990) novel conception of complexity, founded upon constructivist epistemology. A complex phenomenon is then understood here as one that manifests emergent behaviour that is contra-intuitive and not predictable by conventional linear modelling approaches. To this understanding we wish to add an important qualification, as note by J.L. Le Moigne (ibid.), that the property of complexity emerges in the interaction between the observing-subject (one or several people) and the observed-object (e.g. Eriksson, 2007). Hence, complexity is not understood here as being solely an attribute of an independent external object. This implies that a complex phenomenon, like any phenomenon observed, includes one or several human beings that attempt to make-sense of their perceptions and conceptions. Further, in the cases of two or more observers, who thus constitute the understanding of a complex phenomena, differing perceptions and conceptions may easily emerge of this phenomena, which in turn adds another dimension of complexity: we call it here the subject or the human dimension. These differences of perceptions and conceptions – that may be caused by differences in interests or motives, experiences, knowledge, hold value systems and rationalities, gender, etc. – give rise to patterns of plurality (culture) and power-structures (politics), which thus need to be regarded as inherent parts of the complex phenomena at hand.

2.2 Managing Complexity: a Decision-Making Model We rely here on Simon’s (1976, 1982) widely accepted conception of human decision process. This states that decision-making process comprises three main activities of mind, which are Intelligence, Design and Choice. Further, implementation and evaluation often follow these, yet are disregarded in this context. The

204


Intelligence-phase implies the very conception of a distinction between one or several is-situation(s) and one or several ought-situation(s). This implies that there are some inherent criteria (norms) of what ought to be. Secondly, the Design-phase implies the creative act of constructing one or several alternatives for bridging the gap established in the Intelligence-phase, i.e. means of solving the conceived problem. Thirdly, there is the Choice-phase that selects the most suitable alternative for solving the problem. This is in accordance with the previously established criteria. Following Simon, Le Moigne (1990) articulated the cognitive loops of decision-making processes that are exercised by the Choice function. The following is a short account. The decision-making loops are exercised if the Choice function has been unsuccessful in selecting a suitable design option. This can be for various reasons, such as a complex and multi-criteria decision did not generate a satisfying solution. If such an inability of making a choice emerges, the Choice function can activate the Re-Design loop. This feeds back from Choice to Design, where the latter is supposed to generate some new and alternative options. After such a re-design act it is again the function of Choice to attempt a selection. This process of activating the Re-Design loop may be done several times until an adequate option is selected. However, if such an option may not be found due to several Re-Design loops, then the ReFinalisation loop may be activated. This loop feeds back from the Choice phase to the Intelligence-phase. The latter is asked to re-conceive the problem, that is, the definition of the gap between is and ought situations. After such a re-conception the Deign-phase exercise its task and the Choice-phase attempts a selection again. As with the Re-Design process, the process of Re-Finalisation may be looped several times until a successful solution is selected. If however, the Re-Finalisation loop also fails then there is the possibility of activating the Re-Justification loop. This implies a re-conception of the very processor that exercises the conception in the Intelligence phase. This may imply that the basic criteria, which are the various assumptions and values, that ground conception of the is--ought-to gap are to be challenged and modified. In conclusion, the three cognitive loops of decision-making, together with the three phases, make it possible to exercise decision-making. This model of decision-making processes may be understood in the light of the insights into complexity, as identified above. As the key starting function of all decision-making is establishment of the gap between the Is and the Ought-to, such establishment in complex situations may be merely partial, local, and temporal, as unpredictable situations will continuously emerge and hence challenge and transform any established notion of the particular problem at hand. This is also valid for the criteria to be employed for the generation and selection of the solutions. Further, the subject-systems characteristics will impose various understandings and rationalities, sometimes adaptable and sometimes conflicting and coercive. While the expression decision-making may easily put us the limited understanding that a decision is made by one decision-making actor. In complex situations, this is an exception rarely experienced, rather the decisions are polycentric, in that they are conceived and elaborated by a group of actors, that determine this process as discussed above – and whose operations may be re-defined by the mentioned re-justification loop. Finally, the position assumed here is that a decision-oriented IS for complex situations need to support all the three decision-phases, and all the cognitive decisional loops, not only some parts of all these functions.

3. A MODEL OF INTERMEDIATION OF DECISION-MAKING While the discussion above has mainly articulated some key needs for the support of complex decisionmaking as well as some typical limitations inherent in the conception of today’s IS, this second part of the text introduces a model of ‘Intermediation of Decision-Making’. The purpose of this model is to provide an alternative articulation of the decision-making situations, and thereby to serve as a guide for the derivation of guidelines for the conception if decision-oriented IS.

3.1 The Constituents of the Intermediation of Decision-Making As accounted above, a decision-making process implies some kind of gap between one or several issituation(s) and ought-to-situation(s), this is here regarded as the object-system of the decision-making process, or as the problem-space (e.g. Newell & Simon, 1976). Secondly, one or several decision-makers, here called the subject-system, conduct the decision-making process. Thirdly, both the object-system and the subject-system of the decision-making process cannot exist in a vacuum; they are located in an environment,

205

ISBN: 978-972-8924-82-9 © 2009 IADIS

typically an organisation, whether business or other. These distinctions provide constituents of the hereproposed model of Intermediation of Decision-Making. The interactions of these three components are here understood as intermediations, which suggest the need and role for decision-oriented IS.

3.1.1 Conception of the Problem-Space Within their respective organisations, people – i.e. here the subject-system – frequently speak about real problems as if they have actually experienced them. At the same time many of these people, and particularly the strategic decision-makers, have never actually experienced the concrete problem-space, other than perhaps partly. We know that very few senior executives – and sadly also middle mangers – actually meet the customers, or converse with the shareholders, or produce the products, or compile the budgets, etc. These managers typically experience various representations that account for some abstractions of the mentioned, as mediated by an IS, which makes us remember of Baudrillard’s (1981) provocation: ”Territory no longer precedes the map. Instead, it is the map that precedes the territory”. The here-proposed model suggests two viewpoints for problem-space representation: the individual or subject-system, and the environmental or organisational; these two are discussed below.

3.1.2 Intermediation between the Decision-Maker and the Problem-Space A variety of IS-types have been conceived serving the decision-maker information of the problem-space. In general this is conducted by providing a representation of the current situations and also offering various capabilities to investigate alternative potential situations, by means of what-if analyses, extrapolation and simulations, hence the intermediation between the decision-maker and the problem space. In this way, the decision-maker is provided a gestalt of the problem-space and then also creates its alternatives, which is central for the management of complex decision-making. Hence, these IS:s do not only re-present the problem-space, they may also transform and construct it, as noted by Simon (1992). This capability of formulating and construction of alternative current and alternative future behaviours of the problem-space is a key necessity to make unpredictable behaviour intelligible. In this context, our experience is that most of IS do not provide much, if any, functionality for the decision-makers’ to purposefully challenge the relevancy and the validity of the information provided. Also, the type of representations provided, for the articulation of the problem-space, may be crucial – as one cannot learn how to swim from a textbook – there is a need for more diverse types for symbolism and signals. Turning our attention to the other side of this problem-subject intermediation, to the subject-system, or the decision-actors, reminds us that while complex decision making requires most frequently a team of actors, most of IS – of what ever kind – are conceived to serve the single decision-maker, to support their individual articulation, representation, interpretation, and manipulation. This is so, while complex decisionmaking typically imposes upon us a set of interacting decision-makers. These interactions give rise to cultural and political structures, resulting in alternative and conflicting rationalities and political conflicts. And even if all decision-makers of an organisation would utilise one IS, at most they would improve their own cognitive schemata from the decision-maker in charge that has imposed his or her cognitive schemata. However, these remaining decision-actors cannot openly and on similar grounds expose their respective cognitive schemas and values to each other, nor to openly confront these with each other and transform into new formations, problem-spaces and solution alternatives. In our viewpoint, this is a reason why most of today IS:s are not able to support complex decision-making.

3.1.3 Intermediation between the Organisation and the Problem-Space At first, the actions of an organisation are mediated through people’s minds. That is to say, every actor may draw a personal interpretation from his or her representation of the problem-space. But the organisational actions may also be mediated through the information provided by the IS that constitute the organisation (e.g. Sotto, 1997). As one of the key functions of an IS is to store data, this gives the opportunity to build a variety of alternative and complementary representations of the organisational structures and activities. Therefore, an organisational IS can be regarded as a repository of organisational models, i.e. organisational representations, stored for the purpose of some anticipated future use. Once stored and collectively accessible with the help of an IS, these organisational models can be considered as organisational referents, so helpful to co-ordinate organisational actions (e.g. Lorino, 1995) and more specifically, to support process of decision-making.

206


However, such tools seem to be rather permanent or static, which comes from the organisations’ need to establish a constant way of structuring and acting – some would say, the same strategy – wherever it is established. It comes from a classical need for integration of the organisation (Nonaka 1994). In dynamic situations – such as complex decision-making – it may appear necessary for the organisation to define some permanent ways of managing and thus representing, to avoid being “taken in the stream”, which is precisely what our insights into complexity suggest. Rather, this stability becomes a constraint when the situation becomes complex. Radical unpredictability limits the validity and hence the pertinence of such models, which constrain the organisations’ ability to face and manage new situations, or to discover new conditions.

3.1.4 Intermediation between the Decision-Maker and the Organisation The mediation between one or several decision-maker(s) and the organisation as such, shows a slight paradox, as human actors, within the organisation, operate or act as both decision-maker(s) and the organisation. This means that in practice we cannot separate the abstract constructions, or roles, of an ‘organisation’ and a decision-maker from its constituting human actor(s). Further, in some cases one and the same human actor can act both at the same time. The implication for a decision-oriented IS suggests that the intermediation between the decision-makers and the organisation implies communication of information between human actors in the given roles, sometimes dynamic and changing. While communication between human actors as such, within an organisation, has been around since the very upcoming of the first organisations – whether oral, internal papers and memos, telephones, and today e-mails – the context of such an organisation and decision-process, its particular meaning in this context, and also the particular roles of the actors in this context, are not accounted explicitly, and hence subject to confusion. The above discussed bi-polar intermediations do account partially only for the complexity a decisionmaking hence provide only partial comprehension of the decision-oriented IS. It is the task we take on in the next, by focusing the total or global intermediation, where all three constituting components are regarded together and at the same time, with the purpose to suggest some new implications for the conception of decision-oriented IS.

3.1.5. Some Stipulations for the Conception of a Decision Information System The total intermediation accounts for all the three components discussed above, at the same time. Starting with the decision-makers, a decision-oriented IS should support the acting and interplay of a set of decisionmaking actors, rather than one only. This should include the capability to articulate and represent, in various ways, both the individual and the collective cognitive and affective schemata. Further, such an IS should make it possible to support the various transformation and syntheses of these individual and collective accounts. The articulations should make it possible to account for both the perceived and conceived problemspace(s), but also for the decision-maker(s) themselves as well as the problem-space environments, here organisations. The various differences in perspectives and decision-making power need to be articulated and purposefully addressed. Moving our attention to the problem-space, as represented by the set of decision-making actors just mentioned, each such involved decision-making actor needs the ability to articulate one particular problemspace at hand, as perceived and/or conceived, in a variety of versions and alternatives, including conflicting with each other. Further, each such an individual articulation of the problem-space should be possible to articulate in terms of a variety of perspectives or aspects, whether physical, economic, social, legal, ethical, or other, at the same time. All these need relevant symbolism for representation, to justify both the complexity of the problem-space and the perception needs of the decision-maker(s). The decision-making actors and the problem-space interact with its environment, the organisations. Here, two key abilities are needed from a decision-oriented IS. Firstly, the active representation of the organisation as such, for instance in terms of its outputs, their consumers and/or customers, the interaction-channels, the processes, the policies and rules, the various inputs and their suppliers, the variety of their resources, whether human and competence, organisational strictures and cultures, and then physical and informational resources; and the organisational goals. This articulation needs to be dynamic and both account directly any transformations as well as be directed such transformations. Secondly, the organisation needs capabilities for communication between the various human and non-human actors, in their specific roles and contexts.

207

ISBN: 978-972-8924-82-9 © 2009 IADIS

4. SUMMARY The investigation presented here is driven by the question: what conditions should guide the design of decision-oriented IS in order to support the decision-making process in complex human affairs? This question, in turn, is motivated by the observation of the general tendency of today’s IS as informationoriented rather than decision-oriented. The employed approach to answering the question posed is twofold: a conceptual elaboration and exploration of theories of complexity and decision-making and thereby identification of implications for the conception of a decision-oriented IS, and a deconstruction of the decision-making process into its three constituents followed by an elaboration of their intermediations. The results presented here are intentionally explorative, suggestive, and somewhat speculative, and thus in need for rigorous testing, validation and re-formulation. The following list summarises the findings achieved here: - elaborated execution of the intelligence-phase of decision-making; e.g. (a) perceptions and conceptions of the is--ought-to gaps, (b) alternative, polycentric, dynamic and not pre-demanded sources of information collection; (c) alternative and rich representations of the problem-space; - elaborated execution of the design-phase of decision-making; e.g.(a) multiple generation of alternative and conflict solutions, (b) various methods of information manipulation, not only limited to deterministic and stochastic methods, (c) alternative and rich representations of the problem-space; - elaborated execution of the choice-phase of decision-making; e.g.(a) multi-criteria evaluation, that is dynamic; (b) activation of the three cognitive loops of decision-making, the re-design, the re-finalisation, and the re-justification; - elaborated execution of active multi-decision-maker situations; e.g. (a) accounting and articulation the variety cognitive and affective schemata, individual and collective, and their transformations; (b) accounting for the pluralism and political structures inherent in the decision-making team, (c) provide subject-adaptable representations; - elaborated execution of the intermediation of three constituents of the decision-making process, the problem-space at hand, its decision-makers, and the organisation that is its environment.

REFERENCES Baudrillard, J. (1981) Simulacres et Simulation. Paris: Editions Galilée. Eriksson, D.M. (2007). Phenomeno-Semantic Complexity: A proposal for an alternative notion of complexity as a foundation for the management of complexity in human affairs. In: Emergence: Complexity & Organization, Vol. 9, Nos. 1-2. Fulk, J., DeSanctis, G. (1995), Electronic Communication and Changing Organizational forms, Organization Science, 6, 4, 337-349. Holland, J. (1998). Emergence. Oxford University Press, Oxford Kauffman, S.A. (1993). The origin of order. Oxford Univ. Press. NY. Lawrence, P. R. & Lorsh, J. W. (1967), Organization and Environment, Harvard University Press. Le Moigne J.L. (1990). La modélisation des systèmes complexes. Dunod, Paris. Lorino Ph. (1995), Comptes et récits de la performanc:Essai sur le pilotage de l’entreprise. Les Éditions d’Organisation. Newell, A. & Simon, H. A. (1976), Computer Sciences as an Empirical Inquiry. Symbol and Search. Communications of the Association of Computing Machinery, 19, 3. Nonaka, I. (1994), A Dynamic Theory of Organizational Knowledge Creation, Organization Science, 5, 1, 15-37. Simon H.A. (1976). The New Science of Management Decision. Harper & Row, London. 2nd rev. ed. Simon H.A. (1982), Models of Bounded Rationality, Vol. 1 & 2, MIT Press. Simon, H.A. (1992). On designing information for companies and managements in an electronic age. Proc. of the 1992 CEMIT Conf., Tokyo, Sep. Simon, H. A. (1997). The Future of Information Systems, Annals of Operations Research, 71, 3-14. Sotto, R. (1997), The Virtual Organisation, Accounting, Management and Information Technologies, 7, 1, 37-51. Weaver,W (1948), Science and Complexity, American Scientist, 36, 536.

208


3D VIRTUAL WORLD CREATES NEW GENRE OF CYBER SEX D. Yvette Wohn Harvard University, ALM in Journalism 52 Brattle St. Cambridge, MA 02139, USA

ABSTRACT Sex is one of the biggest industries in the virtual world of Second Life—ranging from shops selling beds embedded with animation scripts to professional escort services. While virtual sex is not a new concept, the usage of avatars in a 3D immersive environment creates a level of interactivity unprecedented in former forms of cyber sex. In these virtual worlds, cyber sex includes dialogue as well as visual stimulation, and is unique in that the user is able to actively participate in the content development of both visual and textual elements of role-playing. KEYWORDS Cyber sex, Second Life, virtual relationships, role-playing, virtual worlds, virtual sex, user-generated content

1. INTRODUCTION The Internet has fostered a culture that enables people to engage in sexual activities by using the computer as a medium. Past studies—mainly conducted by psychologists and sociologists—have examined cyber sex as a phenomenon that replaced written text and phone sex (Young, 2001). Cyber sex was defined as messages exchanged in live chat rooms and email correspondence (Dryer, 2007). The recent development of virtual worlds, however, adds a whole new tier to cyber relationships. I am using the term “virtual worlds” to specifically refer to cyber environments that are three-dimensional, such as World of Warcraft or Second Life. The architecture of these 3D worlds enables users to engage in immersive role-playing, which involves text, visual, and, in many cases, audio. By acting out sexual relationships with their avatar—a three dimensional animated figure—users of these virtual worlds can engage in intimate, graphic play that is very different from the existing concept of cyber sex, which is heavily based in text. This paper is based on an independent 18-month study conducted within Second Life as a journalist. This study included observation of commercial areas such as shops, night clubs, and escort services, and separate interviews with 10 individuals who already had experience with sex in Second Life. Due to the relatively small number of Second Life users, I cannot compare the impact of sex in Second Life on society with that of other forms of cyber sex, but am pointing out that virtual worlds have created a new form of cyber sex that not only has sociological and psychological impact on its users, but also plays a large part in the economy and real estate market of virtual worlds.

2. PROLOGUE His wife was asleep when Janus, a 37-year old Web company executive, logged into Second Life at 1 a.m. Waiting for him online was Miso, the avatar—or cyber character—of a 27-year old woman working in the fashion industry. A chat log saved by M reveals a two-hour meeting at a public nude beach starting out with a back massage and leading to multiple intercourse. The transcript shows that the couple had sex in the bedroom of a public beach house, and then two threesomes, where they were joined by random men passing by. “Being in the Internet industry, I’ve heard of people blurring the line [between online and offline] and I used to laugh at them. This totally blew me away. I had no idea it would feel this real,” Janus said.

209

ISBN: 978-972-8924-82-9 © 2009 IADIS

2.1 About Second Life Second Life is a 3D web service where the host company, Linden Lab, gives you an empty piece of land to do whatever you want on it. Second Life users create everything that exists in that virtual world—from shopping malls and dance clubs to esquire service hotels and investment banks. Major corporations such as Sony Entertainment and Reebok use Second Life to promote their products; Cisco uses the system for inhouse communication. Hundreds of schools such as Harvard University and Tufts University use the platform to conduct online interactive education; museums host virtual galleries, and religious and activist groups hold campaigns. Anyone can create an avatar and roam around Second Life for free, but money can buy perks. One can customize one’s avatar in greater detail through clothes, poses, skins, and other accessories that can be bought at user-created stores. A monthly fee (ranging from $6 to $10) enables one to own land, giving one the opportunity of running a business or building a private house. Although many individual users in Second Life spend money for pure leisure, there are also serious business people, such as Ailin Graef, a Second Life real estate developer whose virtual assets equal more than $1 million in real life (Hof, 2006). Among the diverse operations that flourish in Second Life, sex has become one of the biggest industries. Animation designers, sex toy makers, and sexy clothing designers contribute to a multi-million Linden dollar business. People can buy this virtual currency, or cyber money, with real money that trades on an exchange rate that fluctuates depending on supply and demand. Vice versa, money earned in Second Life can also be converted into real-life currency.

2.2 Sex in Second Life Sex in Second Life is conducted by clicking on certain objects that are pre-programmed, or “built,” by other Second Life users. For instance, a bed, couch, or bathtub, will contain an animated script that the user can click on to activate his or her avatar. The user can then choose one or a series of animations. Some objects are programmed for self-satisfaction, such as masturbation or pole dancing, while others require two avatars to get the animation going. During the animation, the user can click on his or her avatar to change positions and choose which avatar gets to initiate the role-playing. At the same, the user can also engage in text or voice chatting.

Figure 1. Sex-related items for sale in a Second Life boutique. Left, an office desk and chair scripted with sex animations. Right, sexy lingerie.

Since these animations are developed by Second Life users based on an open-source code, the quality and quantity of the animations have greatly increased in the past couple years. The types of sexual acts that are portrayed within Second Life are just as diverse as they are in real life—ranging from cuddling to hard-core. Competition also brings down prices, forcing programmers (script writers) to improve on their products.

2.3 How does it Work? A user can choose to purchase an object that contains these animated scripts or go to a location in Second Life that has these objects. Some public areas, such as night clubs or beaches, contain many objects that can

210


be used by anyone for free. For more privacy, players can rent out elaborate villas that come with sex animation-embedded beds.

Figure 2. A screenshot of a male and female avatar having sex in Second Life

Iridium is one Second Life-based business that owns more than two dozen of these villas, charging by the day or the week. The villas are locked and have security features so that only designated people can access the grounds. Users can choose from a number of different house styles, such as a log cabin nestled in snowy forests, or a medieval castle decked out with roaring fireplaces and steaming marble bathtubs. In these houses, bathtubs, carpets, sofas, beds, showers, and other objects that can potentially be linked to sex will contain different sex animations.

3. RELATIONSHIPS BASED ON CYBER SEX Like existing studies on cyber sex—which do not include those in 3D virtual worlds—many of the sexual relationships in Second Life stemmed from problems in real life. With Janus and Miso, for example, it was because they did not have normal offline relationships. Miso, still single, said that she had not had a boyfriend for three years and was too busy with her career to get involved in something serious. Janus, married of ten years, said that he did not remember the last time he had sex with his wife. Both insisted that their relationship wasn’t intentional. “We didn’t jump into sex. It came gradually, like any real relationship. J was my first. I suppose you could say I was a cyber virgin,” Miso said. Not all cases, however, fell into this category. Jazmina, a 31-year old woman who owns a business in Second Life, had a virtual affair with another avatar. Although the affair remained entirely online, it nevertheless jeopardized her marriage. “My hubby was extremely upset when he found out. It didn’t matter that the affair was only cyber. It happens just like real life—you work and play with people and sparks start,” Jazmina said. She said she felt more remorse because her husband had always been very supportive of her Second Life activities. “My business was paying the bills, and he was very proud of me before he discovered the affair bit. It took us counseling to straighten things out,” she said. Then of course, there are some who consider cyber sex as a game. Thoem, a homemaker in her mid 40s, has a male avatar in Second Life. For more than a year, she had a female cyber partner with whom she exchanged eternity rings for an official Second Life partnership, but broke up that love at the expense of 25 Linden dollars to start a new one—with a male partner. The two are part of a large gay community within Second Life and have a master-servant relationship in which Thoem plays a slender, muscled avatar with black bangs and a hairless chest. In several chat sessions held in Second Life, Thoem explained that she doesn’t know her cyber partner’s real-life gender, nor does she care. She is not thinking of disrupting up her real-life family, which includes a 15-year old daughter. At least she can be sure she is not “playing” with her own daughter; minors are only allowed on the Teen Grid—a platform that does not allow any R-rated activities. “It’s an adult version of playing make-believe,” she said. Tabatha, a 28-year old living in Massachusetts, says she is experiencing the pros of a virtual relationship that has lasted for almost two years. Tabatha’s concept of dating, however, does not extend to the real world.

211

ISBN: 978-972-8924-82-9 © 2009 IADIS

So far, the two have only exchanged photographs through e-mail, telephone calls, and of course, meeting in Second Life. She says that they are planning to meet in real life soon and that until now it was all about getting to know each other intimately. “We live far apart, but our personalities clicked and we decided that continuing an online relationship would be more economical,” she said. “And you get to know a person for who they truly are. People tend to feel more comfortable and open when they are behind the computer screen, so they are more relaxed and just being themselves.”

4. OFFLINE IMPACT Second Life is not the only 3D world to be harboring romantic relationships (Scott, 2007), but it is one of the most explicit. Although romantic relationships can take place in online games (Terdiman, 2006), Second Life allows users to engage their cyber characters in actual sex. Watching pornography online is a one-way visual communication while e-mail is two-way communication that is not visual. The combination of interactivity and visual stimulation in Second Life takes cyber sex to a new level. One Second Life user described sex as jumping into a porn video: “Watching the animation is very much like watching porn, except you are the actual character in the porn. It’s a very strange concept for people who have never done it, but the endless possibilities in the virtual world can be an interesting alternative. Of course, you can’t compare real sex to virtual sex. You can’t even masturbate because you’re using your hands to type or click on the mouse. It can be a bit frustrating. Especially when your computer shuts down.”

5. CONCLUSION Despite the multitude of studies conducted of the impact of cyber sex, those studies do not yet include cyber sex in virtual worlds. There is a need, however, for such studies because users claim that sex in Second Life is nothing like online chatting. It would also be interesting to see how voice plays into virtual relationships. Of the people I interviewed, all of the people whose virtual sex led to real life sex had voice conversations with the other person prior to the offline meeting. Regardless of the level of self-representation, perhaps visual and textual elements play a significantly minor role compared to voice. Also, Second Life allows users to visualize their avatars into any form; a user could choose to be human, animal, or machine. With serious communities that focus on role-playing as non-human avatars, Second Life could also be a place to study the behaviors of such people with different sexual orientations. From a technical perspective, cyber sex in Second Life can also be a good example of how open source architecture fuels creativity. Users have taken the basic program offered by Linden Lab and shown great developments in avatar movement, sound incorporation, and avatar animation that requires a fine level of synchronization. Although Second Life is not mainstream, the number of hours users spend on Second Life is rapidly increasing. Recently, Linden Lab, the facilitator of Second Life, reported that user hours were up 61% in 2008 compared to the previous year (Peckham, 2008) which gives us reason to begin observing and discussing how cyber sex is becoming more interactive and immersive.

REFERENCES Dryer, Joy A. et al, 2007. Cyber-sex as Twilight Zone Between Virtual Reality and Virtual Fanasy: Creative Play Space or Destructive Addiction? Psychoanalytic Review, The Hof, Rob, 2006. Second Life’s First Millionaire. Business Week Peckham, Matt, 2009. Second Life User Hours, Dollars Spent, Skyrocket in 2008. PC World Scott, Shavaun, 2007. Email interview with psychotherapist Shavaun Scott Terdiman, Daniel, 2006. .It’s a Nice Day for an Everquest Wedding, CNET news Young, Kimberly S., 2001. Cybersex: Uncovering the Secret World of Internet Sex, Carlton

212


SPECIALIST KNOWLEDGE MANAGEMENT Dr. Mounir Kehal International University of Monaco 2 Avenue Prince Albert II, Principality of Monaco, 98000

ABSTRACT Management practices and information technologies to handle knowledge of satellite manufacturing organizations may prove to be complex. As such knowledge (with its explicit and tacit constituents) is assumed to be one of the main variables whilst a distinguishing factor of such organizations; amidst those specialist in nature, to survive within a marketplace. Their main asset is the knowledge of certain highly imaginative individuals that appear to share a common vision for the continuity of the organization. Satellites and their related services remain a good example of that. From early pioneers to modern day satellite manufacturing firms, one can see a large amount of risk at every stage in the development of a satellite or a related service, from inception to design phase, from design to delivery, from lessons learnt from failures to those learnt from successes, and from revisions to design and development of successful missions. In their groundbreaking book The Knowledge Creating Company (1995), Nonaka et al laid out a model of how organizational knowledge is created through four conversion processes, being from: tacit to explicit (externalization), explicit to tacit (internalization), tacit to tacit (socialization), and explicit to explicit (combination). Key to this model is the authors’ assertion that none are individually sufficient. All must be present to fuel one another. However, such knowledge creation and diffusion was thought to have manifested and only applied within large organizations and conglomerates. Observational and systematic (corpus-based) studies – through analysis of specialist text, can support research in knowledge management. Since text could be assumed to portray a trace of knowledge. In this paper we are to show how knowledge diffuses in a specific environment, and thus could be modeled by specialist text. That is dealing with the satellite manufacturing domain, and having embedded within the knowledge about the business sector and knowledge domain. KEYWORDS Knowledge Management, Corpus-based Analysis, Satellite Manufacturing, Knowledge Diffusion, Text Analysis, SmallMedium Enterprise

1. KNOWLEDGE DIFFUSION IN SPECIALIST DOMAIN In order to investigate the gap in knowledge diffusion within an organization we did carry an observational study, within an SME (Small to Medium Enterprise) in satellite manufacturing, a specialist domain. Inline with a study of the language used in satellite engineering in general, and that stemming from SSTL (Surrey Satellite Technology Limited) and Surrey Space Centre in particular. Both studies have an empirical basis. The observational study (mainly questionnaire-based) was designed to ask questions related to knowledge diffusion within the company during 2002-2005 period, as part of my doctoral research coverage. The questionnaire-based studies were not based on intuitions on how knowledge is managed, rather based on a set of empirical questions, partitioned under five sections namely: Awareness and Commitment, External Environment, Information Technology, Knowledge Maintenance and Protection, and Organizational Issues. We have investigated the diffusion of knowledge within SSTL, based on the practice within SSTL, as articulated through the questionnaire. There were two sets of questionnaire-based observations. The pilot study was conducted with managers and whereas the second run of the questionnaire was intranet-based, and more widespread. SSTL, is a small knowledge-based organization, for minimalism, a knowledge based organisation is one where knowledge is being the dependent input variable, as the need would exist for organisational resources to acquire such knowledge from physical entities (i.e. knowledge workers) and convert it as input for electronic storage medium (s), making it easier for retrieval and dissemination of information. SSTL’s principal assets are its engineers, its project managers and its researchers. Collectively, the engineers, managers and researchers are sometimes called knowledge creating crew (Nonaka et al, 1995). In a rapidly developing, high-technology field like satellite engineering, it is important to communicate, share

213

ISBN: 978-972-8924-82-9 © 2009 IADIS

and validate knowledge. We aim to describe in this paper our understanding of the nature of a specialist organization in a quantifiable manner, and the constructs of a knowledge management audit conducted through the observational study within a satellite manufacturing SME, based in the UK. We have examined how knowledge flows and is adapted between commercial and research types of corpora. One of the major results deduced from the observational study was that knowledge diffusion is paramount within the lifetime of an organization, and could be supported by information systems. Leading us to investigate on how knowledge diffusion takes place, in an empirical way. Our analysis shows that research papers (created within educational institution) and commercial documents (created within spin-offs of such higher education institution) can be distinguished rather on the basis of single word and compound terms. These two specialist lexis show the potential for identifying points of mutual interest in the diffusion of knowledge from the research institution to the commercialization process, thus to application(s) within a domain.

2. METHOD Nonaka et al’s (1995) knowledge conversion model is intuitive. It is based on long experience and judgement. Such model emphasizes the importance of practice, knowledge amongst knowledge workers. The case studies produced were between researchers, practitioners, and managers. There was transfer of knowledge from researchers to knowledge workers. Such has yielded a contingency table for the transfer of knowledge, so-called knowledge conversion model that generates four knowledge conversion modes. Such model is plausible but remains largely intuitive. Our interest is tacit to explicit knowledge conversion (externalization) and explicit to explicit knowledge conversion (combination). The reason we have studied an SME (Small to Medium Enterprise) because it would appear that knowledge would be shared because smaller groups would get together easily, i.e. no logistics involved. As well as it appears that in a SME knowledge bottlenecks which are characteristic of large organizations would not exist. Being in relation to the size of SMEs, managers are expected to interact with and understand needs and requirements of knowledge workers. Consider an organization like SSTL, Surrey Satellite Technology Limited, we focused on the interaction between knowledge engineers and knowledge practitioners, and were aiming to see how knowledge is shared. In order to investigate the gap in knowledge diffusion within SSTL, we did an observational study, and a study of language used in satellite engineering in general. Both studies have an empirical basis. A bimodal research method was followed within the specialist domain of satellite manufacturing applied within SMEs [Small to Medium Enterprise]. Inclusive of Observational study: questionnaire and interview based and Corpus-based study: analysis of text repositories. Involving extraction and modelling of specialist terminology collated from: public domain publications (i.e. NASA, British Standards Institute – Terminology Specification, and BMP - Best Manufacturing Consortium database), and specialist domain publications (i.e. Surrey Space Centre and SSTL). Knowledge Adaptation Observational Study Knowledge workers

Observational Study

Organization Historical Study

Worldwide

Historical Study

Knowledge Flow Figure 1. Knowledge diffusion through flow and adaptation processes

214


Figure 1 represents a relational view of the methodology, integrated within the possible set of agents for knowledge diffusion, being composed of a 2-tier process. Whereby knowledge is assumed to flow among or across from knowledge workers, to the organization, then to worldwide (horizontally), but the adaptation phase comes into place once knowledge is personalised and applied (vertically). However, such methodology was implemented in the specialist nature of the domain of investigation. Yin (1994) identified five components of research design that are important for case studies: the study's questions, its propositions, its unit (s) of analysis, the logic linking the data to the propositions, the criteria for interpreting the findings. The above components were integrated within the observational study, as guidelines to the formulation of the different stages involved within the conduct of this research, from the pilot run of the survey study, to the intranet-based survey and historical studies. In which the intranet-based survey seemed to generate interest, impact and a set of internal actions. Supported as well by over 30% participation rate for the intranet-survey, and being composed mostly by middle to senior managers. Evidence of knowledge diffusion and support for it was manifested as outlined in the table below. Table 1 shows how knowledge bottlenecks have been looked upon during the flow and adaptation of knowledge amid agents involved in its diffusion. Through the observational study and the historical study, behaviour between agents was modelled using the techniques prescribed below. Table 1. Knowledge diffusion in the environment of a small organization Agent A

Agent B

Artefact

Technique

Person

Person

Opinion, practice, know-how, organizational structures

Questionnaire-based study

Person

Organization

PhD Dissertation, Research Publications, technical reports

Text Analysis

Organization

Organization

Specialist documents (i.e. technical documents, technology-

Text Analysis

specific documents, missions documents) Organization

Worldwide

Specialist documents (i.e. technical documents, technology-

Text Analysis

specific documents, missions documents)

The intranet-based survey study had 5 sections, stretching to cover possible areas of applications and implications for knowledge diffusion within an organization, namely - Knowledge Management Awareness and Commitment, External Environment, Information Technology, Knowledge Maintenance and Protection, and Organizational Issues.

3. OBSERVATIONAL STUDY The term knowledge management is used to articulate the concept that knowledge is an asset on a par with the tangible assets of any organisation - land, capital, plant and machinery. Management involves the management of assets; ergo knowledge should be managed from its inception through its nurturing to maturity to exploitation and to ultimate obsolescence. The term was also coined to indicate that knowledge within organisations is communicated not only through the typical organisational hierarchies but also through interaction between members of the organisations across the hierarchies and the different structures (divisions/departments and their functions, management style, communication culture, computer-mediated processes, practices and so forth) contained with an organization. The questionnaire portrays through its five sections, some of the concepts raised within the Knowledge Management field, outlined in Section 1. Two runs for the questionnaire-based study were conducted, a pilot study, and an intranet-based study. The majority of the respondents were knowledge practitioners (i.e. team members). Over 80% from the intranetbased questionnaire were as such, like reported from the respective representative of the study onsite, head of Research and Development at SSTL. The key point was that the managers were more optimistic and confident about extent of knowledge sharing. Our analysis has been supported by the feedback received from

215

ISBN: 978-972-8924-82-9 © 2009 IADIS

one of the key managers cited previously. Our method is no more then holding a mirror to an organization and what is reflected is the management of knowledge within the organization when looked upon from the five different facets of the questionnaire sections (i.e. awareness and commitment, external environment, information technology, knowledge maintenance and protection, and organizational issues). The questionnaire study raises the need for a knowledge map through both the pilot and intranet-based observational studies, one that is specialist in nature. That can represent the domain language providing an environment for querying and validation for the knowledge worker, and thus containment of both elements of such knowledge (explicit and tacit). Allowing as well for the knowledge conversion modes (Nonaka, 1995) to take place, and hence knowledge to be created and utilized. This may act as basis for the research conducted on whether SMEs do create the dynamics of innovation, as such dynamics may need to encapsulate the sharing of the domain knowledge (touted and supported by knowledge workers), and thus embedded within the domain’s ontology – referring to the explicit formal specifications of the terms in the domain and relations among them (Gruber, 1993). This part of the research (observational/introspective) has focused on the organizational structures (management hierarchies, attribution and validation of knowledge, and so forth) in place, enabling or facilitating the diffusion of knowledge. Our conclusions from this survey; based on the feedback and responses received, affirm that knowledge sharing is encouraged. As well as innovation being encouraged either through collective or individual effort(s), and facilitating knowledge sharing is possible through availability of knowledge maps and communication channels between multi disciplinary teams for specialist areas. The above results, from either the pilot study or the intranet-based studies; have encouraged us to explore how a collection of specialist documents will facilitate knowledge diffusion and perhaps to construct knowledge map.

4. TEXT ANALYSIS AND CORPUS-BASED STUDIES The tool used for the purposes of this research is System Quirk, a computational linguistic software system providing a computer-mediated environment for text analysis created by the Artificial Intelligence Group, University of Surrey. The compound terms generated through System QUIRK/Ferret (Artificial Intelligence Group, University of Surrey), illustrate to a certain extent the composition and acceptance of frequent specialist words within a text repository. As well within the language of the domain and the domain knowledge; since latent clusters of concepts, may be represented by each of the compound terms (a hierarchy of concepts through morphological productivity of terms). Whereas each term’s relevance to a collection of documents is erratic, it may be validated (combined and/or externalized knowledge) by the knowledge worker as it composes toward a given terminology. That signifies being of use to the individual knowledge worker or group of them. Thus, achieving acceptance based on consensus within an organization, and growing to be part of it and its external environment (ontological spectrum). We have used these and other sets of compound frequent terms extracted based on a statistical criterion (in relation to the BNC); for comparative purposes, sometimes referred to as “Lexical Signatures” to index collections of text to be contained within a text repository. Frequent compound terms extracted from the collection of documents (Source: Surrey Space Centre corpus) listed below, illustrate the specialist nature of the organization and the domain of the knowledge within aspects of satellite technology encompassed by the research and possibly of the commercial activities of the organization. The compound terms selected below, from the corpus of SSTL and Surrey Space Centre is from a listing of over 50.000 compound terms within the corpus. Such contains all collated research publications of the organization (s) aforementioned. Table 3. Ranking of select compound terms in SSTL/Surrey Space Centre corpora Rank 1 4

Compound Term low cost Propulsion system

Rank 32 33

Compound Term doppler shift swath width

Rank 35 35

Compound Term satellite engineering radiation environment

The above table is illustrative of the morphological productivity (Bauere, 2001) of single word terms, like: cost, satellite, system, launch, sensing, et cetera. Whereas, their compound word formations may be representative of a morphological process based on which knowledge of the domain flows and adapts to the organizational setting in which it is created. Sometimes similar terms that were ranked differently have appeared within the collection of documents collated from the Swedish Space Corp satellite technology news

216


corpus, as shown below in table 4. Being possibly illustrative of wider ontological spectrum (knowledge sharing), of the knowledge of the satellite technology domain and corresponding research and commercial activities. That could as well happen to be dependent on the source of authorship; thus biased. Implying in turn a wider epistemological spectrum (knowledge theory), as suggested by Nonaka et al (1995). Table 4. Ranking of select compound terms in Swedish Space Corp corpora Rank

Compound Term

Rank

Compound Term

Rank

Compound Term

2 3

launch vehicle geostationary orbit

30 31

geostationary satellite launch vehicle satellite launches

49 50

spy satellite(s) satellite launched

The observational and historical studies carried out, have provided better understanding into the field of investigation. Such studies provided the basis and validation for inferences made. Based on Nonaka et al’s (1995) terminology used within the knowledge conversion model, portraying creation of knowledge and corresponding conversion processes. It is believed that knowledge undergoes a combination and socialization conversion process (for knowledge flow) within an organization or across a (sub) domain (s), and undergoes an internalization and externalization conversion process (for knowledge adaptation) within an organization or across a (sub) domain (s). Flow of knowledge (Combination & Socialization of knowledge) Research Documents Research Documents Research Documents Research Documents Research Documents (1…n) systematic diffusion Adaptation of knowledge (Internalization & Externalization of knowledge) Commercial Documents Commercial Documents Commercial Documents (1…n) chaotic diffusion

Figure 2. Knowledge diffusion through the knowledge conversion model (Nonaka et al, 1995)

5. CONCLUSIVE REMARKS

The case study is a method of learning about a complex instance through extensive description and contextual analysis. The product is an articulation of why the instance occurred as it did, and what may be important to explore in similar situations, in our case the specialist knowledge and its diffusion is the product. As the observational study laid the framework for the conduct of our research, it was focused on examining knowledge flow, and corresponding practices and information technology support in place. Results from the observational study have indicated that knowledge bottlenecks may exist, in particular were technological support could be needed. The transmutation of science into technology is a complex process when one sees unique ideas highlighting the past scientific landscape and beneficial technological artefacts in the present. The notion of satellite technology or space technology, with variable scope and scale, was an ostentatious idea that has led to a range of remote sensing and earth observation instruments for instance. The unique idea is a key reference point for forecasting how the idea will metamorphose into an artefact. Knowledge is communicated through so-called semiotic systems: written text, images, mathematical and chemical symbols, and so on. The knowledge of emergent domains is yet to standardize its symbol systems which simply add to the (creative) chaos inherent in such emergent systems. The analysis of change in written text, amongst the

217

ISBN: 978-972-8924-82-9 © 2009 IADIS

most changeable semiotic system at the lexical level at least, may reveal a consensus or dissension in the use of terms. Terms denote concepts and textually help us to understand how knowledge evolves in an emergent domain. The emergent domain of small satellite technology was studied as an exemplar. This is our attempt to establish a method, which covers a broad range of texts, research articles, commercially-driven documents and state-of-the-art papers representative of research and development conduced within an organization, to observe the emergence of a new domain. We have by design focused on an innovative organization to establish our method which is driven by knowledge workers, document-based and guided by terminology utilized. The method will facilitate the construction of knowledge maps in an objective and systematic fashion. It is an intuitive statement that research ideas and experimentation form the basis of new technologies, products, and practices. The research effort leads to the creation of new knowledge, and to the suspension of ‘obsolete’ knowledge, and this knowledge crosses over into technology. Perhaps a comparative analysis of the choice of terms (lexical signature) will indicate the extent of this cross-over. Our analysis shows that research papers and commercial documents can be distinguished somewhat on the basis of single word and compound terms that were generated automatically. These two lexical signatures show the potential for identifying cross-over points in the diffusion of knowledge from the research arena to applications domain. The metamorphosis of science into technology is a complex process when one sees innovative ideas highlighting the past scientific landscape (i.e. in the form of PhD theses and state-of-the-art research papers) and beneficial technological artefacts in the present. The notion of satellite technology, with variable applications, was a unique idea that has led to a range of remote sensing devices for example. The innovative idea is a key reference point for forecasting how the idea will metamorphose into an artefact. Knowledge is communicated through so-called semiotic systems: written text, images, mathematical and chemical symbols, multimedia and so on. The knowledge of emergent domains is yet to standardize their symbol system which simply adds to the (creative) chaos inherent in such emergent systems. The analysis of change in written text, amongst the most changeable semiotic system at the lexical level at least, may reveal a consensus or dissension in the use of terms. This is our attempt to establish a method, which covers a broad range of texts, PhD theses, journal articles, technical reports, and state-of-the-art review papers, to observe the emergence of a domain and hence specialist diffusion of knowledge.

REFERENCES Atkins, S. Clear, J. Ostler, N. 1992. “Corpus Design Criteria”. Literary and Linguistic Computing, 7(1), 1-16. Bauere, L. 2001. “Morphological productivity”. Cambridge: Cambridge University Press, (Cambridge studies in linguistics; 95). Biber, D. Conrad, S. Reppen, R. 2002. “Corpus Linguistics: investigating language structure and use”. Cambridge: Cambridge University Press. Burrows, J. F. 1992. “Computers and the Study of Literature”, pp: 167-204. Oxford: Blackwell. Gruber, T.R. 1993. “A Translation Approach to Portable Ontology Specification”. Knowledge Acquisition, Vol. 5, 199– 220. Holmes, D. I. 1998. “The Evolution of Stylometry in Humanities Scholarship”. Literary and Linguistic Computing, 13: 111-17. Liebowitz, J. 2000. “Building organisational intelligence: A knowledge management primer”, Boca Raton, CRC Press. Nonaka, I. Takeuchi, K. 1995. “The Knowledge Creating Company: How Japanese Companies create the Dynamics of Innovation”. Oxford University Press, Oxford. Sinclair, J. 1991. “Corpus, Concordance, Collocation“. Oxford, Oxford University Press. Sperberg-McQueen, C. M. 1991. “Text in the Electronic Age: Textual Study and Text Encoding, with Examples from Medieval Texts”. Literary and Linguistic Computing, 6/1: 34-46. Stemler, S. 2001. “An Overview of Content Analysis”. Practical Assessment, Research & Evaluation, 7(17). http://ericae.net/pare/getvn.asp?v=7&n=17. A peer-reviewed electronic journal. ISSN 1531-7714. Last Accessed [15/11/2002]. Surrey Space Centre (University of Surrey) – SSTL (Surrey Satellite Technology Limited), George Edwards Library (University of Surrey), collection of text documents 1979-2002, http://www.surrey.ac.uk/Library, http://www.ee.surrey.ac.uk/SSC/, www.sstl.co.uk. Swedish Space Corp, collection of text documents, historical corpus 1997-2002. http://www.ssc.se. Yin, R. 1994. “Case study research: Design and methods”. (2nd ed.). Beverly Hills, CA: Sage Publishing.

218

Reflection Papers


DOVETAILING WITH DIVERSITY: A CHALLENGE FOR HCI Sebastiano Bagnara Faculty of Architecture, University of Sassari, Alghero, Italy

Simone Pozzi Dept. of Communication Sciences, University of Siena, Italy & Deep Blue, Research and Consulting, Rome, Italy

ABSTRACT In this paper, we maintain that disabilities should be considered as a particular case of human-technology integration, by focusing on the need for HCI to study diversity rather than disabilities. We reflect on the different perspective on disability adopted by World Health Organisation in 2001 and on how it should inform the relationship between HCI and disability. Designing for disability is part of more general challenges, i.e. designing for diversity, that HCI is facing primarily because of the digital revolution and subsequent changes in our lifestyle. To tackle this issue, HCI should not consider disability as a niche, as an isolated problem to be addressed only by some special dedicated means. We should instead frame HCI applications for disability in the overall context of changes in the nature of human-machine interactions. KEYWORDS User-centred design, ergonomics, human-computer interaction, disability, accessibility.

1. INTRODUCTION The field of HCI studies on disability has definitively certain peculiarities. HCI researchers do need to study the nature and origins of specific physical or cognitive conditions, in order to design user-friendly technologies. But it also shares the basic goal of the general HCI “Human-computer interaction is a discipline concerned with the design, evaluation and implementation of interactive computing systems for human use and with the study of major phenomena surrounding them” (Association for Computing Machinery, 2008). Our paper reflects on the relationship between HCI and disability. We propose to consider disability not as an isolated field of HCI study. It is rather part of a general challenge, that HCI is facing as a discipline in its all of its applications (a similar claim can be found in Lewis, 2006). Our claim is that HCI is not yet fully equipped to cope with the increased complexity and variety of our technological life, of which disability is only a specific part.

2. DOVETAILING AND INTERNALISATION At the core of HCI studies is indeed the special relationship that we as humans are able to establish with technological devices. As well put by Andy Clark (Clark, 2003), humans are increasingly becoming “naturalborn cyborgs”. Key concept in his reasoning is the one of ‘dovetail’. By interacting with tools and external elements, we put our cognitive skills in connection with these external elements, to form a new unity, new cognitive skills, new abilities, where it is hard (and not very meaningful) to distinguish between the contribution of our mind and that of external tools. “Ours are essentially the brains of natural-born cyborgs, ever-eager to dovetail their activity to the increasingly complex technological envelopes in which they develop, mature, and operate” (pg. 26). Behind such an approach lies a theoretical tenet that has gained momentum since the eighties, as a reaction to the cognitive revolution. Its main claim is that cognition is not only in our head, but results from

221

ISBN: 978-972-8924-82-9 © 2009 IADIS

the connection between our head and the external world, in particular tools. As a product of their functioning, our brains internalise tools. Our tools become part of our brain. Brain plasticity is not something we have to try hard for, it is like second nature. And it applies to any tool.

3. CHANGES IN THE NATURE OF WORK HCI is now in the middle of the so-called information revolution, by the end of Tayloristic era. Indeed, most of today work is very far from that of twenty-five years ago. At that time, industrial work prevailed. It was very simple, usually performed in dedicated place, permanent, and repetitive. It did not require mental effort, and was predominantly manual. Clerical work was similar, even though implied routinary mental, rather manual work. The actual prevailing form of work shows different characteristics (National Research Council, 1999, Malone, 2004). From the demographical point of view work is heterogeneous. The diversity in human resources is growing in terms of gender, race, education, culture and status. Work is becoming more dynamic and boundaries among various jobs have become weak and permeable. Many workers are engaged in more parallel tasks, sometimes exploiting different competencies. Interruptions and distractions are the fabric of such a multi-tasking work, driving away from reflection to action. However, new work forms tend to rely on the same technological infrastructure. Compared to the Tayloristic era, today’s workplaces are more similar, all implying the use of a personal computer. Working tools are now part of our houses and of our personal life: technology is everywhere, and employers expect people to have the skills required to manage this technology. Those who lack these skills risk being cut off from work. This digital divide is not limited to work but is also present in everyday life and affects people’s social relations.

4. DISABILITY AS LIMITATION OF ACTIVITY This fundamental transformation clearly affects also the inclusion of disabled people at work. And it has not gone unnoticed by the World Health Organisation (WHO), that in twenty years has changed the basic definition of disability. In June 2001, the WHO published “Classification of Functioning, Disability and Health”. The title is indicative of substantial changes. The aim was no more to describe the handicap, but the state of wellbeing of a person, by focusing on her/his own physical state, but also on the different ways that a person interacts with the outside world, and the impact of external events on a person. There is no reference to any disorder, structural or functional, without linking it to the state of ‘wellbeing’. Therefore the WHO classification includes not only the physiological and cognitive functions and the physical structure of a person (the nervous, visual and auditory systems, the vocal apparatus, the cardiovascular and respiratory systems, etc.), but also the activities that guarantee inclusion and social participation (learning, communication, interaction and interpersonal relations, community life, etc.), and environmental, i.e. natural (the environment), artificial (technology), and social (support, attitude, services, etc.) factors. The classification system covers every aspect of human health, arranging them in two different domains. The first is the health domain, which includes the action of seeing, hearing, walking, learning and remembering, etc. This is directly related to physical structure and functions. The second, the health-related domain which includes mobility, education, participation in social life and similar, refers to activities and environmental and social factors. The WHO definition does not only concern persons with disability but everybody: it has a universal use and value. The classification is ‘positive’. It starts from the ‘normal’ level of functioning, considers if a person differs from this norm, and how far they are from it. The term ‘handicap’ is abandoned and the term ‘disability’ is extended to cover the limitation of activity and restriction on participation. For the function and structure of the body, the qualifier can assume values from zero (no impairment) to four (severe impairment, equal to 96-100%). Similar qualifiers exist for the activities that do not refer to impairment, but to limitations, and for participation it is said that there are restrictions. In short, concerning environmental factors there are barriers. It appears that new WHO definitions and classifications are strongly related with the changes in work (not to say in the attitudes of disabled people toward them) that had had taken place during the years. Moreover,

222


all this has been strongly influenced by the information technology. In summary, handicap is now seen as a form of diversity in a society where all are diverse and need some kind of assistance.

5. HCI AS SCIENCE OF DIVERSITY So it is work itself that demands for an increased diversity, because it has become more complex, more technology enmeshed, because it mixes the traditionally separated moments and spaces of work life and personal life. Disabilities have to be seen as some of the forms of diversity we should take care of, since, as for technologies, difficulties and benefits concern everyone. It depends on the tool, and the context. Difficulties can arise or change at any time; it is not possible to overcome them once and for all. This perspective implies the concept of disability must be replaced by the idea of diversity. From the point of view of diversity, people have not to be qualified as disabled, but rather as more or less diverse from others with respect to their working situation and the degree of their IT competence. Everyone is, in a way, disabled with respect to digital technology. To overcome their personal digital divide (subjective and contextual), everyone needs to personalise technology. The needs of disabled people are cases in the general necessity to personalise IT. Disabled people may live in greater symbiosis with technology, so they are probably the most evident case of human-machine symbiosis. The challenge for HCI is to turn itself into a science of diversity, that no longer mainly focuses on adapting the machine to the human. This transition has not been completed yet, even though it is acquiring increasing importance in the HCI community (Harper et al., 2008). In this last part of our paper, we will briefly outline three main open issues that, in our opinion, HCI has to address to turn itself into a science of diversity: (i) From work to self-realisation, (ii) A new pace of learning, (iii) Digital ecosystems

5.1 From Work to Self-Realisation As a consequence of the wide-spread use of computers outside workplaces, HCI needs to address not only work-related problems, it should also cover leisure and entertainment (Bodker and Sundblad, 2008). HCI researches now need to go beyond ‘human needs’ and focus on ‘desires’. People nowadays engage in interactions with the machines not only to work, but most of all for the experience of it, to fulfil personal goals. HCI is not only about correcting interaction problems, it has to create fulfilling experiences. The same reasoning should apply to HCI and disabilities. HCI should not be primarily focused on solving impairments, it should also include in its objectives the creation of rewarding experiences of use. In a recent interview, Donald Norman aptly pointed out that assistive devices should be also aesthetically pleasing (“CNN Designers challenged to include disabled”, available on the Web on Norman’s site www.jnd.org/). For instance, social networks are valuable not because of their usability, but because they serve as attracting poles for a community of people. This source of value does not change if an user interacts with it with the aid of assistive technologies.

5.2 A New Pace of Learning A second challenge for HCI comes from the current pace of change. We are not only referring to technological innovation. We are mostly concerned with the rate of change of human desires. If HCI had to correct interaction problems, we may expect these problems to be fairly stable. That is why we have design heuristics and principles. Desires are instead by their very nature dynamically changing, as they are referred to the whole of a person, including social relationships, values, culture, one’s own lifestyle. One person may even have contradictory desires, subject to drastic changes over time. The line of reasoning for disability should be similar. Users need to learn how to adapt to assistive devices’ rate of change. HCI has to adapt its methods and techniques to such a pace, possibly supporting users in the continuous learning activity required by innovation. Disable people see the continuous change in the technologies, in the way they have to interact with them, in the skills and knowledge needed to operate with them, as requiring an endless effort to cope with (Bagnara and Failla, 2007).

223

ISBN: 978-972-8924-82-9 © 2009 IADIS

5.3 Digital Ecosystems Some HCI researchers maintain that “a good tool is an invisible tool” (Weiser, 1994). This may be true in many respects, but what about the perceived degree of control over technologies? What about troubleshooting activities? How can we form our model of the system functioning, if the system works almost independently from our inputs? Both degree of control and troubleshooting rely to a large extent on conscious analysis and reflection. If computers are invisibly weaved into our world and if they start to invisibly interact with each other, the user is likely to have little (if any) visibility on their behaviour. The challenge for HCI is to analyse and design human-computer interactions, computer-computer interactions, but most of all the degree of control that users should have on these digital ecosystems. We would need to design new interfaces to enable control of digital ecosystems, to represent their state in a comprehensible manner and support intervention. Disability and assistive technologies are again a “not that special” case of the above dynamics. Assistive technologies often form a deeper symbiosis with their users, so HCI need to pay a special attention to loss of control and invisibility. Our claim here is not that we should avoid designing for the embodied interaction, rather that HCI should devise methods and techniques to explore the tensions between experiential and reflexive interaction, to find means to identify the right trade-off.

6. HCI AND HUMAN VALUES HCI is facing the challenge of an ever increasing diversity. HCI application fields are not stable at all. This means an enlarged scope for HCI research. It also means including human values in HCI leading principles. Technologies are not neutral, they have an impact on socio-cultural values. Such an impact is probably the only sensible criterion to steer design. HCI should be able to analyse the multi-faceted issues brought by innovation, be those of a cultural, social, or even ethical nature. HCI should progress from technology-driven, to user-centred, to human-centred (Harper et al., 2008). Design trade-off are likely to increasingly concern socio-cultural aspects, with large variations across different communities (Medhi, 2007). We should address more dynamic aspects like desires, moving from mass-market products to multitudemarket products. We no longer study the interaction between technology and an undifferentiated user (that has to be defined as broadly as possible to fit anybody), rather we need tools to address every user’s needs and diversity. Even very peculiar ones. Design to overcome handicaps has no more to be seen as a niche, but rather as the most challenging endeavour of widespread, general approach: dovetailing with diversity to design human-centred innovations. Limitation of activity and restriction on participation are two key concepts not only for disabilities, but also for HCI in general. They are not only related to physical, cognitive or social impairments, but also strongly due to the changing nature of society, work and IT use.

REFERENCES Association for Computing Machinery, 2008, Definition of HCI. Bagnara, S. & Failla, A., 2007, From Handicap to Diversity. Lecture Notes in Computer Science. 613. Bodker, S. & Sundblad, Y., 2008. Usability and interaction design-new challenges for the Scandinavian tradition. Behaviour and Information Technology, 27, 293-300. Clark, A., 2003. Natural-Born Cyborgs: Minds, Technologies, and the Future of Human Intelligence. Oxford University Press, Oxford, UK. Harper, R., Rodden, T., Rogers, Y. & Sellen, A. (Eds.), 2008. Being Human: Human-Computer Interaction in the year 2020. Microsoft Research Ltd, Cambridge, UK. Lewis, C., 2006. HCI and cognitive disabilities. interactions, 13, 14-15. Malone, T. W., 2004. The Future of Work: How the New Order of Business Will Shape Your Organization, Your Management Style, and Your Life. Harvard Business School Press, Cambridge, MA. Medhi, I., 2007. User-Centered Design for Development. interactions, 14, 12-14. National Research Council, 1999, The changing nature of work: Implications for occupational analysis. Washington, DC, National Academy of Sciences. Weiser, M., 1994. The world is not a desktop. interactions, 1, 7-8.

224


ROLE OF ELECTRONIC RECRUITMENT Seppo Sirkemaa Turku School of Economics, Pori Unity P.O. Box 170 Pori – Finland

ABSTRACT Internet is increasingly being used in recruiting personnel. Electronic media allows new ways to find qualified staff. There are differences in these services, especially when it comes to recruitment services offered by individual companies. Here we explore electronic recruitment methods in larger organizations. We found interesting differences among the researched organizations in the available recruiting services. KEYWORDS Recruitment, electronic services, e-recruitment

1. INTRODUCTION Employment is a challenging issue in our world. Paradoxically, at the same time we suffer from unemployment and lack of competent employees. Here the focus is on organizations which are seeking qualified and motivated employees to work for them. The challenge is to find the right person to each vacant task. Modern technology, especially Internet can help in matching the workforce with the employers. In the old days it was customary to go to the employment office in order to find a job. Compared to this Internet brings benefits to people looking for work. It makes it easy to rapidly find vacancies and get information on the organizations looking for employees – conveniently with the computer. Electronic recruitment gives new opportunities to faster find right person to the right job. It may not eliminate the need for traditional ways of recruitment – but can significantly speed up the process and give more control to both the employers and employees. Let us look closer at these issues.

2. RECRUITMENT IN THE DIGITAL AGE Recruitment is about finding workforce, qualified employees so that company can target towards the operative and strategic goals. The processes within the organization cannot be optimized if there are no qualified employees (Roberts, 2002). As a result, finding the right person to the right task is the goal of recruitment. In addition, also employee's opinion of organizations work environment, future career opportunities etc. impact the recruitment process and it's successfulness.

2.1 The Recruitment Process The recruitment needs vary in organizations. There are also differences in the way recruitment needs are recognized, employees are evaluated, decisions are being made on which candidate to select etc. As a result, the recruitment processes in organizations are different. Recruitment practices that are well suited for one organization may not fit others. This is the case also in electronic recruitment services. Despite organizational differences in recruitment practices there are general stages that are typical to all recruitment processes. These include (Markkanen, 2002)

225

ISBN: 978-972-8924-82-9 © 2009 IADIS

planning – where the need for additional staff is being recognized and required competencies that the new employee should have will be mapped. This stage includes also decisions like what channel and media would be optimal in order to find candidates for the job. - implementation, managing applications – The candidates send applications and the organization will decide which candidates will be taken into consideration. The process of managing and evaluating application sis critical in this stage. - implementation, interview – getting to know the candidates better - decision – evaluating the candidates based on selected criteria and choosing the most suitable candidate to the job (Roberts, 2002) - follow-up – evaluating the successfulness of the recruitment. Tutoring, and discussions with the new employee. Here the target is to integrate the new employee to the organization and improve the recruitment process in the organization Thorough planning and understanding the criteria which will be used in evaluating and choosing among candidates are cornerstones of successful recruitment. This is also the case in electronic recruitment process. The application form is usually the method used in mapping candidates suitability for the job. It may be used to identify key issues in recruitment like skills, educational background, previous career and motivational aspects (adapted from Roberts, 2002). One of the most important functions – at least from the employee's perspective – is that it is used as a basis for deciding which candidates will be eliminated at this point and which ones will be asked to the interview (Wood and Payne, 2000). -

2.2 E-recruitment The digital revolution and the Internet have affected also the recruitment process. For companies moving recruitment services to Internet seems appealing: instead of advertising in newspapers, for example, potential employees can be met on-line on a 24/4-basis. Generally, recruitment is one of the areas where Internet can have a major impact – both traditional recruitment and technology empowered electronic recruitment or shortly e-recruitment are based on information about available jobs (Galanaki, 2002). In this paper is e-recruitment is defined as recruitment that takes advantage on the Internet. It includes information on available work and the company that offers it, but includes also interaction and transactions that are connected to recruitment. Interactive elements allow the employee to fill an application and send it right away. These are here referred to as e-recruitment applications. (Galanaki, 2002; Markkanen, 2002). How can companies take advantage of e-recruitment? There are three main options here (Galanaki, 2002): - adding information about vacant jobs and recruitment in their company to existing web-pages (Scheyer and McCarter, 1998). This can be a very cost-effective way to deploy e-recruitment - simultaneous advertisement in printed media and internet. For example, the same add can be in a newspapers and on newspaper's website (CIPD, 2006) - using portals and web-sites which specialize in recruitment related services. These can include listings on vacant jobs, registers and CV's, for example (Rudich, 2000) It must be noticed that all approaches to e-recruitment may not fit all companies. Smaller companies may benefit from using portals and specialized web-sites in seeking employees (Hansen, 1998). On the other hand, well known companies can benefit from a variety of e-recruitment approaches (Greengard, 1998). The decision on how to take advantage of e-recruitment is therefore a company-specific challenge

3. EMPIRICAL PART In the empirical part we are focusing on the e-recruitment services that companies have made available for people looking for work. The questions of interest are related to the type e-recruitment services that are available, and we wanted to study what is the added value of these services to the person who is looking for work. We approach these questions from the employees perspective, and do not for example look at how electronic recruitment services have been developed in the organization. From the potential employee's viewpoint available information and transaction-related features of the web-site are important in order to make the employee interested in working in the organization.

226


In this research we studied the e-recruitment services by looking at the web-sites of all companies that employ 50 or more in Satakunta, an area is situated in the western coastal part of Finland. It was assumed that larger organizations employ more often than manager-owned smaller companies. Processes that are connected to recruitment are therefore important for successful recruitment in the selected companies. Based on the above mentioned selection criteria the group of case organizations consists of 137 companies. About 25 percent of these companies employ more than 250 persons. At the same time, two third of these companies have an annual turnover over 10 million euros. The e-recruitment services are being evaluated based on the e-service framework presented by Ancarani (2005). There does not exist a framework for e-recruitment, and therefore we have modified the stage-based framework of e-service sophistication (Ancarani, 2005) and included also selected viewpoints from the eservice diamond framework (Goldkuhl and Persson, 2006). As a result, the web-sites of case companies are classified into: level 0: there is general information on the company level 1: the web-site contains information on recruitment opportunities in the company level 2: there is a job-application form on the web-site level 3: web-based application form which can be filled in and sent right away level 4: there is a enhanced web-based application form which can be filled in and sent The first three levels in this classification are rather straightforward, they are modified from the e-service framework as presented by Ancarani (2005). The distinction between forms on level 2 and higher levels is that on level 2 the form is in electronic format but it has to be downloaded, filled in separately and then sent either electronically or by traditional mail for processing. On level 3 and 4 the service includes a form which can be filled in and sent for processing without the need for downloading forms etc. However, it can be questioned how can these levels be identified, specifically what makes an e-recruitment application “enhanced”. An enhanced application contains following features: - there are multiple choices or drop-down menus in at least three places in the application - the potential employees competency is studied by having fields for education, language skills, earlier work experience, IT skills or other relevant skills(at least three fields) - there are various attributes in the application, for example when asking about work experience there could be fields for company, task and time when instead of a single field for asking about earlier work experience - the application consists of several pages or screens, one has to fill in a screen in order to open the next one - one can add fields when needed, for example new fields can be added so that all work experience can be listed (no fixed number of fields or space) - it is possible to attach files such as a picture or a curriculum vitae The e-recruitment service is classified as highest level, enhanced web-based application when at least three enhanced features are present in the application.

4. CONCLUSION The contribution of this paper is in increasing information on different recruitment methods. Here we have studied e-recruitment and argue that it brings added value to the recruitment process. From the employees perspective internet and e-recruitment have dramatically changed the process of finding vacant jobs and employment opportunities (Jansen et al., 2005). E-recruitment makes it possible to contact the employing company directly, and the potential employee can get to know the company before the applying to the company. In fact, companies should take into consideration what is on the web-site: if the information is not relevant to the job-seeker he or she might head to more interesting sites. The impact that web-site has on the image is significant, and it does affect also recruitment. Most of the researched companies give information about the company, field of operations etc. but all did not have e-recruitment services available through the internet. The company size seemed to be an important factor here, as only in larger companies there were more interaction and transaction oriented e-recruitment. Small companies had basically only information about the company on their web-sites.

227

ISBN: 978-972-8924-82-9 © 2009 IADIS

Today people who are looking for work may find information from companies’ web-sites which helps in finding the right job. Internet-users are becoming more familiar with e-recruitment in their everyday life, and so it becomes natural to use internet also in looking for a job. As a result, companies wanting to find the best employees might want to develop their e-recruitment services, or other companies pick the best of them.

REFERENCES Ancarani, A. (2005). Towards quality e-service in the public sector: The evolution of web sites in the local public service sector, Managing Service Quality, Vol 15., No. 1, pp.6-23. CIPD (2006). Chartered Institute of Personnel and Development. Online recruitment. , retrieved 14.3.2007. Galanaki, E. (2002). The decision to recruit online: a descriptive study. Career Development International, 7 / 4 (2002), pp. 243-251. Greengard, S. (1998). Putting online recruitment to work. Workforce, Vol. 77, No. 8, 7377. Hansen, K. (1998). Cybercruiting changes HR. HR Focus, Vol. 75, No. 9, 1316. Jansen, Bernard J., Jansen, Karen J. and Spink, Amanda (2005). Using the web to look for work – Implications for online job seeking and recruiting. Internet Research, Vol. 15, No. 1, pp. 49-66. Markkanen, M. (2002). Onnistu rekrytoinnissa. WS Bookwell Oy: Juva. Roberts, G. (2002). Recruitment and selection. The Cromwell Press: Wiltshire. Rudich, J. (2000). Job hunting on the web. Linkup, Vol. 17, No. 2, 2124. Scheyer, R. and McCarter, J. (1998). The Employer’s guide to Recruiting on the Internet. Impact Publications, USA. Trocchia, Philip J. and Janda, Swinder (2003). How do consumers evaluate Internet retail service quality? Journal of Services Marketing, Vol. 17, No. 3/2003, pp. 243-253. Wood, R. and Payne, T. (2000). Competency Based Recruitment and Selection. A Practical Guide. John Wiley & Sons Ltd: West Sussex, England. Yang, Z., Jun, M. and Peterson, R.T. (2004). Measuring customer perceived online service quality – Scale development and managerial implications. International Journal of Operations & Production Management, Vol. 24, No. 11/2004, pp. 1149-1174.

228


TOWARDS IMPROVED CONCEPTS OF APPROPRIATE USABILITY Barrett S. Caldwell, PhD Purdue University West Lafayette, IN 47907 USA

ABSTRACT In many situations, the goal of the human factors and ergonomics professional is to eliminate the complexity and thus increase the usability of devices and interfaces. While information and communication technology (ICT) systems should not be so complex as to prevent users from effective participation in the Information Society, there are reasons to avoid too strong an emphasis on simplifying at all costs. This paper addresses potential costs and dangers of making ICT interfaces too simple, and considers an emphasis on achieving an effective balance of complexity and usability to improve users’ ongoing involvement in ICT development and social integration. KEYWORDS Cognitive skill; user interface design; usability; ICT; information society

1. INTRODUCTION Since the introduction of increasingly complex information and communication technology (ICT) devices and interfaces, there have been complaints about the excessive difficulty of using such tools. It is certainly true that television remote controls, cellular telephones or digital cameras with hundreds of features (and user manuals of hundreds of pages) can often defeat the purpose of improved access to technology. Thus, the human factors and ergonomics profession has developed a strong reputation for attempting to assist users and improve ICT designs by reducing the complexity of ICT devices and therefore increasing their usability. Is it possible, however, to overemphasize the push for increasing usability? This paper addresses the question of whether there is a danger to making ICT devices and interfaces too easy or simplified for effective use. Although it may seem at first glance to be a positive effort to make ICT interfaces as easy as possible, there are at least three reasons why such an effort may result in negative long-term consequences. Each of these reasons encourages us to consider a previously under-examined concept of “requisite complexity ” and “appropriate usability”. This paper addresses each reason from the perspective of improving overall robustness of human participation in the information society.

2. ARGUMENTS AGAINST EXCESSIVE EASE OF USE This paper does begin from the perspective that the goal of effective ergonomic design is to enable and increase the ability of the ICT user to successfully participate in the information society. Such participation is an ongoing process, and not simply an individual act of technology use. In this context, usability must consider a variety of issues regarding the interaction of user, activity, and context factors influencing the use of an ICT device to achieve a goal (Bailey, 1996). The larger social and task context of ICT use should encourage the ergonomist to consider “beyond the screen” or utility elements of why and when the device is used to support the user’s goals, and not simply the “at the screen” or usability elements of activating particular functions (Caldwell & Uang, 1995). The following three considerations represent reasons or arguments to suggest that the simplest possible interface may not be the most supportive of human activity.

229

ISBN: 978-972-8924-82-9 © 2009 IADIS

2.1 Cognitive Fitness and Engagement A growing body of work addressing the plasticity of the brain has begun to suggest that continued engagement in cognitively challenging tasks helps to maintain effective mental functioning throughout the lifespan (Doidge, 2007). There are at least two underlying concepts that support this approach to cognitive fitness. The first is a direct analogue to physical fitness—a person must maintain a reasonable level of activity in order to be considered physically fit. Ongoing participation in cognitively engaging tasks, then, allows the person to continue to develop and maintain neuronal connections that enhance ongoing intellectual capability. From this perspective, it is actually better for the long-term health of the individual to be routinely challenged (but not overwhelmed) by new cognitive activities. Too much ease of use suggests the cognitive equivalent of a “couch potato” changing television channels via remote rather than going outside to engage in physical activity. Of course, a person who goes directly from no physical activity to the attempt to run a marathon will find the task impossible, and likely cause severe physical damage. The cognitive fitness perspective, however, has no clearly defined analogue to physical exercise guidelines. What is the appropriate level of cognitive difficulty and engagement for an ICT device? Current ergonomic research does not provide many clues, but cognitive difficulty in ICT use is a common issue in computer game design (Juul, 2009). In order to be seen as engaging and interesting, a game must be seen as having a level of complexity and difficulty sufficient to give the user a sense of accomplishment when s/he succeeds. The capability for success will also vary with user expertise, so that the game must provide increasing levels of challenge with increasing levels of expertise. In fact, as Juul points out, it is the moderate likelihood of failing (not certain failure, but not guaranteed success) that provides the sense of engagement and satisfaction for game play (Juul, 2009). A similar perspective can be found in the classic work of Murray (1938) describing manifest needs for achievement and mastery, and environmental presses or conditions that allow such needs to be met. The simplistic ICT device or interface that does not allow increased levels of skill or achievement would be seen as a hindrance, rather than a help, to achieving the user’s goals. In addition, this interaction between user needs and ICT support (press) raises the issue of appropriate opportunities for the user to learn the range of ICT functions and capabilities over time.

2.2 Increased Learning of ICT Functions The perspective of learning ICT functions can be seen as having a connection to the previous discussion of engagement and success with computer and video games. However, the ability to explore and interact with an ICT system in order to develop increased expertise with that system has been shown as relevant in a number of other contexts. Exploration of interface functions and modes of ICT system performance is considered a primary mechanism by which operators develop expertise and situation awareness with those systems (Endsley, Bolstad, Jones, & Riley, 2003; Endsley, Bolté, & Jones, 2003; Sheridan, 1992). One form of human-machine interaction error, known as “mode error,” arises from the operator’s failure to understand the current system state, functional range and performance modes of the ICT. There are thus two sources of danger from an interface that does not allow the user sufficient range of interaction and complexity of use. 1) An excessively simplified interface belies the complexity of modes or performance capabilities of an ICT device, preventing the user from developing sufficient awareness of that complexity. 2) A lack of feedback regarding the current mode or system state of the ICT prevents the user from developing an experiential understanding of the human-machine dynamics of the ICT, and thus limiting their development of expertise. As an ICT system becomes more complex, the range of system functions is expected to increase. This increasing range of functions suggests that the user should spend increasing time interacting with the ICT in order to develop expertise and contextual understanding of how, when, and for what the ICT can be effectively utilized. However, ICT systems with too many functions can result in the problem that the user has fewer opportunities to interact with the ICT across conditions to develop a robust understanding of its capabilities, as well as sufficient training experience with the ICT in those conditions. Increasing complexity of ICT devices, combined with decreasing periods of experience, can thus result in technology development cycles that become decoupled from both user and developer feedback cycles and expertise gain processes. This can be considered a potentially damaging phase of diffusion of innovations (Rogers, 1983), where the

230


characteristics of the innovation are no longer well known, well utilized, or well practiced before the next generation of innovation requires a return to novice states of ICT interaction.

2.3 Changing Price/Performance Demands Three main factors help to exacerbate the capability of rapid cycles of ICT diffusion and distribution. The speed of information technology innovation and gains in computing power allow for ICT technologies to gain in the number and variety of features without an appropriate consideration of user requirements and capabilities. It is often this breakdown between technology capability and user experience that increases the calls for ergonomists to work to simplify interfaces. However, a cultural phenomenon has affected the pace of ICT development. While the legacy of Moore’s Law (transistor size and computing power continues to increase at an exponential rate) has continued over a period of decades, companies have historically used the implications of the Law to increase device complexity. The relationships between computing power and device development, however, could be used equally well to provide similar functionality for lower price rather than increasing functionality for similar price ("Less is Moore," 2009). If companies were to apply this model of decreasing price rather than increasing functionality more frequently, there would be additional opportunity to stabilize an effective ICT design and enable a broader diffusion of that design. By pursuing such a strategy, a company may also develop an improved understanding of which ICT interface features were most effectively utilized (or conversely, most poorly integrated into device use), and thus develop a more effective functional integration of features to further improve the price / performance ratio of the device. Instead of adding features that contribute limited value to the device, even further gains in performance can be achieved with increasing device reliability and robustness.

3. CHANGING EXPECTATIONS AS A COST OF INAPPPROPRIATE USABILITY In the mid 1990s, the author and his laboratory considered the issues of excessive dependence on ICT advance based on the context of a relatively specific technology: radio communications for American football players (Caldwell, Everhart, Paradkar, & Suh, 1995). Mass marketed consumer ICT devices, however, add an additional consideration to the concern of excessive dependence on technology developments. If users are led to believe that they do not need to spend time learning or developing expertise with ICT systems that they buy, these users are in fact becoming more detached from the ICT and its influences on the Information Society. In short, the lack of appropriate usability based on mismatches in cognitive fitness, expertise gain, and price-performance ratios can ensure that users are paying more to utilize relatively fewer features of the ICT, at decreasing levels of effectiveness and deep understanding. Over time, a severe disconnect could result from such trends, where large fractions of society become functionally incapable of effectively utilizing the technologies which they acquire, and still call for technology development changes that decrease, rather than increase, their net capability. This eventuality seems a direct contradiction of the preferences and goals of the ergonomist. Unfortunately, the change in expectations can be seen as a technical requirement for the development and sale of the next generation of ICT devices; users may not see the advantage of buying new technologies that promise to do less and require more cognitive involvement. The economic climate of the end of the first decade of the 21st Century, ironically, may provide a stimulus to revise these changing expectations. Manufacturers have been forced to rethink their marketing strategies, and must now consider developing new lines of less expensive products to respond to depressed international demand for consumer products. Longer technology development cycles may result from lower industrial capacity, fewer employees, and slower sales rates. It is no longer profitable for companies to create new generations of consumer ICT products if existing unsold inventories remain that still exceed user requirements. Larger forces are serving to change economic and consumer expectations in novel ways, which may also spur an increased interest in appropriate usability and robust ICT utilization.

231

ISBN: 978-972-8924-82-9 © 2009 IADIS

4. CONCLUSION Despite the primary thesis of this paper, the author does not mean to suggest that the ergonomic search for usability is a misguided one. Excessive complexity to ICT interfaces, and devices that cannot be used as intended due to poor user interaction design, fail to achieve the goal of effective technical design or economic efficiency. However, there are multiple sources of inefficiency that can serve to degrade the long-term effectiveness of ICT systems and their users. Additional research is certainly required to determine what ranges of complexity of interface design are best suited to maintain and enhance user cognitive fitness and awareness of “beyond the screen” aspects of new ICT devices. Optimum usability and utility remains an overarching goal for the ergonomist. However, considerations for the context of ICT use in the Information Society can help to enhance, rather than degrade, the cognitive health and technological fluency of users as they interact with an appropriately designed range of information and communication technology systems.

REFERENCES Bailey, R. W. (1996). Human Performance Engineering: Using human factors / ergonomics to achieve computer system usability (3rd ed.). Englewood Cliffs, NJ: Prentice Hall. Caldwell, B. S., Everhart, N. E., Paradkar, P. V., & Suh, H. (1995). Issues of Dependence and Reliance on Technological Advancement: Examples from American Football and Air Traffic Control. Proceedings of the Human Factors and Ergonomics Society 39th Annual Meeting, San Diego, CA. Caldwell, B. S., and Uang, S.-T. (1995). Technology Usability and Utility Issues in a State Government Voice Mail Evaluation Survey. Human Factors, 37(2), 305-323. Doidge, N. (2007). The Brain that Changes Itself. New York: Penguin Books. Endsley, M. R., Bolstad, C. A., Jones, D. G., & Riley, J. M. (2003). Situation Awareness Oriented Design: From User's Cognitive Requirements to Creating Effective Supporting Technologies. Proceedings of the Human Factors and Ergonomics Society 47th Annual Meeting, Denver. Endsley, M. R., Bolté, B., & Jones, D. G. (2003). Designing for Situation Awareness. Boca Raton, FL: Taylor & Francis. Juul, J. (2009). Fear of Failing? The Many Meanings of Difficulty in Video Games. In M. J. P. Wolf & B. Perron (Eds.), The Video Game Theory Reader 2 (pp. 237-252). New York: Routledge. Less is Moore (2009, January 15, 2009). The Economist. Murray, H. A. (Ed.). (1938). Explorations in Personality. New York: Oxford University Press. Rogers, E. H. (1983). Diffusion of Innovations (3rd ed.). New York: The Free Press. Sheridan, T. B. (1992). Telerobotics, Automation, and Human Supervisory Control. Cambridge, MA: MIT Press.

232


THE INTERNET, OUR CYBORG MIND Nuno Nodin Instituto Superior de Psicologia Aplicada, Lisbon, Portugal

ABSTRACT This theoretical article examines some of the ways by which Humanity has been affected and how it has affected technology. Using the concept of prosthetics and briefly overviewing some key authors’ reflections that illuminate our discussion, this article specifically focuses on how the Internet has been integrated into our thought processes. Our analysis uses Donna Haraway’s metaphor of the cyborg to illustrate this phenomenon. Following up on other authors, we further suggest that human subjectivity is undergoing a shifting process due to its’ coupling with computers and with the Internet, in a process that questions the very notions of self and identity, as well as that of human nature. KEYWORDS Internet, cyborg theory, haraway, technology

1. OF MAN AND TECHNOLOGY It is impossible to conceive of the evolution of Humankind without considering its intimate relationship with technology. More than a collection of artifacts that make life easier, technology has been an active element of humanity’s development through time. In fact, some claim that technology has developed, as Man has, within a symbiotic relationship through which both have gained and grown (Hein, 1991). According to Mark Pesce (1993), new technologies, throughout man’s evolution gave way to new abilities and therefore “introduced a change in the content and structure of human culture. To state that these extensions by themselves produced these shifts in culture is to oversimplify; a complex relationship of feedback loops between humans, their artifacts, culture, and systems of communication together create these changes. (op. cit., online) Neil Postman (1994), on a similar note, considered technology to have an ecological impact upon society. For this author, the introduction of a new technology totally changed the social and cultural landscape of the context where it is introduced. Echoing on this view, but applying it to the individual, Pesce (1993) stated that “[a]ny technological amputation [meaning use of a new technology] always has a consequent effect in the structure of the self, as the reconfiguration of the senses produced by self-amputation introduces a new gestalt, or world view” (online). Man-made objects have progressively replaced biological functions with artificial ones, thus alleviating the body of effort and simultaneously extending Man’s abilities. Wooden sticks have always been used to assist the elderly and the physically impaired walking, glasses and contact lenses have been improving people’s sight for a long time, dialysis machines help overcome kidney malfunctioning, and so on and so forth. At same time, telephones, television and webcams have enhanced our communication abilities, shortcircuiting time and space in ways never before imagined. The more advanced the technology, the more sophisticated the functions people are able to perform, thus extending their capacities. Although the notion of prosthetics as applied to the substitution of limbs or teeth for man-made replacements was developed in the 16th century, that date doesn’t actually mark the start of the use of physical add-ons to the people who needed them for medical reasons, as those have been present since ancient times (Coffey, online). Man has therefore been aware for a long time that artifacts could assist overcome physical shortcomings. Consequently, on a conceptual level, many thinkers have endorsed and developed the idea that physical objects and technology can be conceived as extensions of people in many different ways. One of the first thinkers to somehow make that connection was John Locke (1689). In his discussion over property rights, he suggested that by mixing objects with labor these objects gain the same rights as those of the body of the laborer.

233

ISBN: 978-972-8924-82-9 © 2009 IADIS

[E]very man has a property in his own person: this no body has any right to but himself. The labour of his body, and the work of his hands, we may say, are properly his. Whatsoever then he removes out of the state that nature hath provided, and left it in, he hath mixed his labour with, and joined to it something that is his own, and thereby makes it his property. (“Of labour”, Locke, 1689)

Similar suggestions were made by Karl Marx also within the context of his reflection over labor, as well as by Freud in Civilization and its Discontents (1929) where the Viennese doctor wrote that “[t]he idea of men’s receiving an intimation of their connection with the world around them through an intimate feeling which is from the onset directed to that purpose sounds so strange and fits in so badly with the fabric of our psychology that one is justified in attempting to discover a psycho-analytic – that is a generic – explanation of such a feeling” (p. 65). This is what this author sets himself to do in that book. Following up on these particular topics raised by Freud and others, modern scholars have taken a step further and questioned the traditional division between psychological functions and the rest of the world in ways never envisioned before. The idea of an outsourcing mind, for instance, has been put forward by many scholars, challenging the traditional views that the mind is bounded by the skull. Gregory Bateson (1972), for instance, discussed how a blind man’s stick could be considered a functional extension of his mental system, as it is not possible to conceive that man’s psychomotricity without the elements that are an integral part of his walking. More recently, others (Dennett, 1996; Clark & Chalmers, 1998) have also suggested that cognitive processes’ use of external resources in fact proves that cognition is not limited to the brain. If, as we confront some task, a part of the world functions as a process which, were it done in the head, we would have no hesitation in recognizing as part of the cognitive process, then part of the world is (so we claim) part of the cognitive process. (Clark & Chalmers, 1998, p.8).

This theory of mind, named Active Externalism (Clark & Chalmers, 1998) extends the concept of prosthetics to the realm of the psychological. For the purposes of the current paper we would like to underline the idea, shared by these authors, that the boundaries between internal and external mental processes is not as clear as traditionally thought. Notions of prosthetics and of extended minds raise many questions about our bodies and about ourselves. Where does one’s body ends and the outside world commence? What happens to men and women when pieces of the physical world are integrated into the body? Do we become less human and more machine? That is when the concept of the cyborg becomes helpful. The word cyborg was coined by Manfred Clynes and Nathan Kline (1960) in relation to the possible use of self-regulating technological systems to help men survive in space. Five years later, Clynes wrote in the introduction of a book by D. S. Halacy (1965) of a new frontier, that of the “relationship between 'inner space' to 'outer space' – a bridge... between mind and matter” (op. cit., p.7). That is the frontier that the cyborg inhabits. More visible in science fiction than in real life, the half man, half machine creature portrayed in films like “The Terminator” (1984) or “Robocop” (1987), was recycled on a refreshing and inspiring, albeit ironic way, by Donna Haraway in her “Cyborg Manifesto” (1994). She defined the cyborg as a “cybernetic organism, a hybrid of machine and organism, a creature of social reality as well as a creature of fiction” (p.83). Haraway’s project is, in essence, a feminist one. She uses the idea of cyborg as a way to move past binary views of gender and away from seminal Western discourses, such as those of Christianity or Psychoanalysis, that conceive of women as inferior to men. For her, with the cyborg “[t]he dichotomies between mind and body, animal and human, organism and machine, public and private, nature and culture, men and women, primitive and civilized are all in question ideologically.” (p.86). With it, she also dismisses the love-hate relationship that Man has always had with technology. In fact, while machines were seen during the Modern Age inaugurated with the Industrial Revolution as the solution to many of Man’s problems, at the same time fear of alienation from machines has always been present. This has been reflected, for instance, in science fiction films such as “Metropolis” (1927), “2001: A Space Odyssey” (1968), “The Matrix” (1999), or “I, Robot” (2004) where sentient computers and machines control or are used to control and dominate humans. Haraway looks beyond those dystopian fantasies and embraces the machine as part of Man, as a source of pleasure and as an object of desire. In her words, “[t]he machine is not an it to be animated, worshipped, and dominated. The machine is us, our processes, an aspect of our embodiment.” (Haraway, 1994, p.108). Haraway’s strong metaphor of the machine-organism hybrid and her questioning of boundaries has become an inspiring one following up on what others before her suggested. It has been used by many others

234


to think differently about the relationship between Man and technology, not only from a purely physical point of view, such as the one Haraway mainly envisioned, but also from a psychological one. Ruff-Stahl (2005), for instance, used it to explain problem solving while piloting military airplanes. Others have applied it to how the Internet is affecting different dimensions of our selves (Bortle, 2005; Hamman, 1996). The cyborg metaphor does, in fact, fit nicely in explaining the ways cyberspace has been integrated into our mental space. A visionary in his own time, J.C.R. Licklider (1960) put forward the idea of a man-computer symbiosis. For him, "[t]he hope is that, in not too many years, human brains and computing machines will be coupled together very tightly and that the resulting partnership will think as no human brain has ever thought" (Licklider, 1960, p.4). In many ways we can see that symbiosis taking place in our lives today, even without us being physically connected with the Internet, as in the futuristic vision set by William Gibson’s sci-fi novel Neuromancer (1984), or with game consoles as in the film eXistenZ (1999). Set in a not distant future these tales, both of the dystopian type dismissed by Haraway, picture a world where machines plug directly into human brains, thus submersing the user in a new reality that becomes as real as that of the tangible physical world. Another author to discuss media as extension of self, also many years before the wide availability of the Internet, was Marshall McLuhan in particular in his book Understanding Media: Extensions of Man (1964). There he suggested that Humanity has approached the final phase of extensions by or through the technological simulation of consciousness, in particular through television. Today (…) we have extended our central nervous system itself in a global embrace, abolishing both space and time as far as our planet is concerned. Rapidly, we approach the final phase of the extensions of man - the technological simulation of consciousness, when the creative process of knowing will be collectively and corporately extended to the whole of human society, much as we have already extended our senses and nerves by various media. (McLuhan, 1964, p. 3-4)

McLuhan’s thinking about media offers a far wider (and sometimes contradictory) perspective on this topic than is possible to include in this particular discussion. However, his account of how electronic media provides a sort of extended consciousness is easily applicable to the realm of the Internet, an electronically mediated space exponentially larger than television. Some have put forward that indeed “skulls and skin do not bound mental systems, and through computer networking these systems can now extend indefinitely” (Mitchell, 2003, p.36). The concept of the cyborg again comes to mind, a being that not necessarily needs technology incorporated into the body to rightfully become a cyborg because technology is already integrated in the brain. Computer screens, mice and keyboards simply provide the interface where flesh meets machine. The Internet and cyberspace include both dimensions of the cyborg selves we have discussed thus far. If the physical devices that allow us to be connected, such as palm pilots, cell phones and personal computers, can be understood and are often experienced by people as extensions of themselves (Turkle, 2004), the Internet may on several accounts also be conceived as an extension of our psyche. For instance, we now use the web as an extension of our memory, accessing the information it contains in ways we previously used our own mind. Email, instant messaging systems and blogs allow us to contact others much more efficiently than ever before. With it, we continue to do what we were best programmed to do – to communicate with others – but much more and better than by simply using our organs of senses and of communication. By using the Internet we become not necessarily better or worst but certainly different. In the context of what has been discussed so far, one of the most striking questions that we can ask is: What sort of beings are we becoming with this symbiotic relationship with the Internet? If History is made of continuous loops between Man, who builds machines, and machines, who in turn transform the nature of Man, what sort of artifacts do we become in the process? Haraway suggested the rise of a new identity: “[t]he cyborg is a kind of disassembled and reassembled, postmodern collective and personal self.” (Haraway, 1994, p.102), a perspective echoed by Sherry Turkle (1995) when discussing the multiplication of the self in a myriad of online identities. For Mitchell (2003), “it isn’t simply that our sensors and effectors command more territory, that our webs of interconnectivity are larger and more dynamic, or that our cell phones and pagers are always with us; we are experiencing a fundamental shift in subjectivity” (p.62). The question of how we are shifting towards a post-human or inhuman identity (Sim, 2001) may well be dated, as that identity is already us. However, it is one that hasn’t been sufficiently discussed, argued, disrupted or integrated, even as we are living it. That may well derive from the fear of alienation from machines we have carried for great many years, while at the same time we are drawn to them in ways that often defy explanation (Nodin et al., 2008). These questions will be more and more crucial as new

235

ISBN: 978-972-8924-82-9 © 2009 IADIS

technologies emerge and continue to force rethinking the nature of our humanity, while at the same time they will become part of it.

AKNOLEGEMENTS The author would like to thank Prof. Alex Carballo-Diéguez of the HIV Center for Clinical and Behavioral Studies, New York City, USA; Prof. Gary Dowsett of the Australian Research Center in Sex, Health and Society, Melbourne, Australia; and Prof. Isabel Leal of Instituto Superior de Psicologia Aplicada, Lisbon, Portugal, for their ongoing support and help in the development of this manuscript. This work was supported by grant SFRH/BD/17396/2004 of Fundação para a Ciência e a Tecnologia, Portugal.

REFERENCES Bateson, G., 1972. Steps to an ecology of mind; collected essays in anthropology, psychiatry, evolution, and epistemology. Chandler Pub. Co., San Francisco. Bortle, J. J., 2005. Games people play: Identity and relationships in an online role-playing game. Ph.D. dissertation, Duquesne University, United States -- Pennsylvania. Retrieved February 20, 2009, from Dissertations & Theses: Full Text database. (Publication No. AAT 3199507). Clark, A., and Chalmers, D., 1998. The extended mind. Analysis -Oxford-. 58 (1), 7-19. Clynes, M.E. and Kline, N.S., 1960. Cyborgs and space, Astronautics, American Rocket Society, 26–27, 74–75. Coffey, S. (online). Prosthetics. Accessed online on March 9, 2009, at http://csmt.uchicago.edu/glossary2004/prosthetics.htm Dennett, D. (1996). Kinds of minds. New York, NY: Basic Books. Hamman, R.B., 1996. Cyborgasms. Cybersex amongst multiple-selves and Cyborgs in the narrow-bandwidth space of America Online chat rooms. MA Dissertation. Department of Sociology, University of Essex, Colchester, UK. Haraway, D., 1994. A manifesto for cyborgs: Science, Technology, and Socialist-Feminism in the Late Twentieth Century," in Seidman, S. The postmodern turn: new perspectives on social theory. Cambridge, Cambridge University Press. Hein, M., 1991. The erotic ontology of cyberspace. In Benedict, M. (Ed): Cyberspace: first steps. Cambridge, MIT Press. Freud, S., 1962. Civilization and its discontents. W.W. Norton, New York. (original 1929) Gibson, W., 1984. Neuromancer. Ace Books, New York. Horrocks, Chris. 2000. Marshall McLuhan and virtuality. Postmodern encounters. Icon, Cambridge. Halacy, D. S. 1965. Cyborg; evolution of the superman. Harper & Row, New York. Licklider, J. C. R., 1960. Man-Computer Symbiosis. IRE Transactions on Human Factors in Electronics, vol. HFE-1, 411. Locke, J., and Laslett, P., 1988. Two treatises of government. Cambridge texts in the history of political thought. Cambridge University Press, Cambridge [England]. (Original 1689) McLuhan, M., 1964. Understanding media: the extensions of man. Routledge & K. Paul, London. Mitchell, W. J., 2003. Me++: the cyborg self and the networked city. MIT Press, Cambridge, Mass. Nodin, N., Leal, I., Carballo-Diéguez, A., 2008. “Através da máquina é mais fácil” – Definições e reacções ao conceito de cibersexo [“Through the machine it is easier” – definitions and reactions to the concept of cibersex]. Proceedings of Ibero-American Conference IADIS, WWW/Internet 2008, Lisbon, Portugal. 129-135. Pesce, M. ,1993. Final Amputation: Pathogenic Ontology in Cyberspace. Paper presented at the Third International Conference on Cyberspace, University of Texas, Austin, May 1993. Accessed online on October 20, 2008, at http://hyperreal.org/~mpesce/fa.html. Postman, N., 1993. Technopoly: The Surrender of Culture to Technology. Vintage Books, New York. Ruff-Stahl, H.S., 2005. The intuitive cyborg. Problem solving in the post-computational age. Dissertation submitted to the Division of Media and Communications of the European Graduate School in Candidacy for the Degree of Doctor of Philosophy, Berlin. Sim, S., 2001. Lyotard and the inhuman. Postmodern encounters. Icon, Duxford. Turkle, S., 1995. Life on the screen: identity in the age of the Internet. Simon & Schuster, New York. Turkle, S., 2004. Whither Psychoanalysis in Computer Culture?. Psychoanal. Psychol., 21, 16-30.

236


INTERNATIONAL MOBILITY OF WOMEN IN ICT CAREERS Núria Vergés Bosch, Ana María González Ramos, Rachel Palmen Gender and ICT Programme. UOC University

ABSTRACT In recent decades, the movement of people is greater given the intensification of economic activities and the improvement of ICT. The mobility of scientists plays a double role in strengthening professional careers and improving R&D and innovation systems. In the ICT sector, labour supply in skilled jobs is much wider, and although women are much less represented than men, they may adopt different mobility behaviours in comparison to women in other areas of knowledge. In this article, therefore, we ask whether there are gender differences in ICT career mobility and what could help to explain them. In this short paper we aim to briefly introduce the issue of women’s international mobility and present some of the findings of our work in progress about flows of human resources in Science and Technology in relation to gender and Spain. The transformation of traditional gender roles in the developed world and the modernisation process in developing countries has enabled women to adopt a more active role in the process of mobilisation. In Spain, however, deep gender differences are still present. Women still represent a minority in ICT areas of knowledge; therefore they are still underrepresented as mobility grant beneficiaries in these areas. However, the scarcity of women in the ICT area is accompanied by a higher success rate in mobility grants among women than their male counterparts in the same area. According to the literature and our findings, we can conclude that gender is an influential factor of mobility in ICT careers. KEYWORDS Mobility, Gender, ICT, Highly Skilled Human Resources

1. INTRODUCTION The mobility of scientists plays a double role in strengthening professional careers and improving R&D and innovation systems. In the ICT sector, labour supply in skilled jobs is much wider, and although women are much less represented than men, they may adopt different mobility behaviours in comparison to women in other areas of knowledge. In this article, therefore, we ask whether there are gender differences in ICT careers mobility and what could help to explain them. The transformation of traditional gender roles in the developed world and the modernisation process in developing countries has enabled women to adopt a more active role in the process of mobilisation. In Spain, however, deep gender differences are still present. Women still represent a minority in ICT areas of knowledge; therefore they are still underrepresented as mobility grant beneficiaries in these areas. However, the scarcity of women in the ICT area is accompanied, by a higher success rate in mobility grants among women than their male counterparts in the same area. According to the literature and our findings, we can conclude that gender is an influential factor of mobility in ICT careers.

2. INTERNATIONAL MOBILITY OF WOMEN Mobility is defined as a change of residence, temporarily or permanently, for labour, social or personal reasons. In recent decades, the data shows that women from countries with fewer resources are more likely

237

ISBN: 978-972-8924-82-9 © 2009 IADIS

than men of the same nationality to move to other countries. This could be related to different reasons summarized by the following key factors (Skachkova, 2007; OCDE, 2007): • The need for the female population to leave their country of origin to progress professionally. • The difficulties that women have to overcome, for example invisible social and professional barriers in their country of origin. • The abandonment of the traditional role of women. In the past female migration projects were associated with their families’ migration projects, thus related to women’s traditional roles and accompanying their partners. • At present, female mobility is subject to profound social changes, not always separated from family strategies when women are in a relationship. The negotiation of professional careers is a new factor to be analysed which can facilitate the prioritisation of female careers. • In some collectives and nationalities it has been empirically demonstrated that women’s mobility processes, carried out alone or accompanied by other female relatives, are linked precisely to the subsistence of their families living in their country of origin. Although the validity of these hypotheses requires a specific in depth study, we consider that any of these circumstances, either independently or combined, increases the probability of female mobility. These circumstances have reduced the importance of masculine mobility in relation to global mobility. The extension of educational attainment achieved by the female population allows all countries to count on a high percentage of highly skilled female workers. Mobility in tertiary education in non OECD countries indicates that there are higher rates of emigration among women (13,9%) than among men (9,7%). In Africa, the rate increases to 27,7% among women compared to 17,1% among men. In Latin America the percentage is 21,1% for women compared to 17,9% for men. The difference is less significant in Asia and Oceania and there is no gender gap in Europe and Northern America (OECD: 2008a, 2008b). Docquier et al. (2007) found a high correlation between migration rates, gender gap and country of origin schooling rates. According to this data, where the gender gap in education is higher, the female population is more encouraged to seek employment opportunities abroad, promoting higher emigration rates for women. Their study found that the proportion of skilled women migrants has increased from 46.7% to 49.3% in the period of 1990-2000. This increase is due to the increased supply of a qualified female labour force as a result of educational credentials in developing countries. Other studies show similar mobility percentages between women and men, especially early in their careers. This trend changes in the case of older women, when women postpone their mobility, or decide not to move, due to personal or family reasons (Ackers: 2004, 2005). Social pressure can play a negative effect on women’s self esteem and their intention to pursue the professional challenges posed initially. The progressive abandonment of women in various stages of scientific careers is an empirically proven fact, as the gender gap increases more dramatically at the highest positions in comparison to lower levels within the same area of knowledge. In scientific careers it is common to find couples who research in the same area or work in close environments. This phenomenon, known as a dual-career, has been investigated empirically, showing that women are more likely to have partners of the same profession than men. This situation often puts women in a subordinate position with respect to their male partner, because men’s careers are often prioritised. The achievements of men are considered more important by the couple, which is detrimental to the career advancement of women, whose careers become "tied" to the careers of men (Ackers: 2005). Female researchers remain the weaker party in terms of the number of children they have which is clearly evident in statistics. Male scientists are more likely to have a greater number of children than female scientists, who are also more likely to remain unmarried than their male counterparts (Saltford: 2005). Ackers (2004) outlines that while male careers are not subject to any requirement, other than the personal project, female careers are subject to men’s careers, family strategies and the couples decision making, influenced by established social roles and the set up of the labour market (wage levels, expectations of stability and career advancement) Men can develop their professional career consistently with a family life more easily than their female colleagues (Teichler and Mainworm: 1997). Women are more often in a situation of incompatible choices i.e. the decision to devote herself fully to her personal life or to her career. Male colleagues who are married, to a greater extent, have partners outside this demanding environment. That is why it is easier to find male

238


scientists with more children than women scientists. Mobility projects can split up couples. It is easier for a woman to abandon a promising career than a man - doing so in favour of her partner’s career. For women it becomes more costly to pursue a full career if this means their performance as mothers becomes compromised. It is not lightly that many couples consisting of skilled professionals choose to delay the age of procreation until a certain state of stability is reached. This does not happen if less importance is given to the female career or if she has a profession in which she is already established. Therefore, the longer and more demanding the scientific career - regarding the number of hours of work, the more problems women encounter, preventing them from taking up an all consuming career that acquires prestige and success. Moreover, women with dependents experience a lack of freedom of movement which impedes their more frequent mobility, - currently necessary for networking and attending events- that contributes to professional visibility and, ultimately, the achievement of a more advantageous position within the most influential scientific community (Addis, 2004). This behaviour becomes a recurring negative social process. The Mateo effect, is applicable to the situation described, and clarifies the reason why it is more difficult to find women in relevant positions of scientific and professional careers.

3. MOBILITY IN ICT CAREERS IN SPAIN FROM THE GENDER PERSPECTIVE In order to analyse international mobility in ICT careers we have carried out research within the Spanish context. Firstly, we analyse publicly available data in Spain in relation to professional mobility and secondly, we present the main results of the analysis of our qualitative research. As in most countries, participation of women in technology is a minority. Thus, in Spain, women represent approximately 20% of ICT students and professionals (Pérez Sedeño: 2004). The minority presence of women in the ICT areas is also reflected in the number of grants they receive. This horizontal segregation, together with vertical segregation, has to be faced by women in all areas of knowledge to achieve positions of greater responsibility (Castaño: 2008; González Ramos: 2009) Figure 1 shows the unequal distribution of mobility grants between men and women, which is evidence of the existing huge gender gap. In 2004-05 only two out of six women would have received a mobility grant, and even less if we consider only the support granted in the ICT area. In this case, only one in ten people would be women. 2004 Total areas TIC area

Women 32,0 7,3

2005 Men 67,9 92,7

Women 28,5 11,9

Men 71,5 88,0

Source: AGE Figure 1. AGE Mobility grants distribution. Areas of knowledge and Gender (%)

Scarcity of women in the ICT area is accompanied, however, by a higher success rate among women than their male counterparts in the same area, which clearly demonstrates female capacity. This indicator measures the ability to get a scholarship, taking into account the demands and the final number of awards in each program. ICT areas show a high success rate in most of the calls under the Spanish central government, a trend that also remains true in the case of mobility programs. Despite this, in figure 2, the success rate in the ICT area for 2005 is lower due to the decrease in the number of awards this year, as shown in figure 1. 2004

2005

All areas

Women 4,0

Men 2,7

Women 3,3

Men 2,3

ICT area

5,2

2,1

2,0

1,7

Source: AGE Figure 2. Success rate in AGE Mobility Grants. Areas and Gender (%)

239

ISBN: 978-972-8924-82-9 © 2009 IADIS

Figure 2 shows the success rate between the sexes for all areas of knowledge and ICT areas. Women show a higher success rate in all areas, especially in ICT. In Spain there is no quota policy, and the highest percentage of female success must be attributed to the ability of women to obtain scholarships. We must conclude that the women involved in this knowledge area are very well trained. Figure 3 also suggests that the presence of women is somewhat higher in those programs where the age cohort is younger. Second, post-doctoral training programs present a higher proportion of women among the beneficiaries. Female participation among foreign professors and technologists is third, followed by sabbatical and Spanish professors where women represent a minority. 2005 Postdoctoral Talent return Foreigners Spanish Sabbatical

Women 25,0 17,6 11,1 5,0 6,3

Men 75,0 82,4 88,9 95,0 93,8

Source: AGE Figure 3. Women and Men participation in mobility programs managed by AGE in ICT areas, 2005.

These results are consistent with the literature that suggests the lower tendency of women to develop flows of international mobility when reaching a certain development in their career because of their family. Of course, it should also be noted that in ICT areas of knowledge it will also be less frequent to find women in the professional categories of greater responsibility. This gender gap is present not only in Spain, it is common to other European countries and developed countries. So that even in mobility programs for foreign professors and technologists’ gender segregation in ICT is also noticeable. The qualitative analysis that has been conducted in our research has allowed us to delve into the reasons for the reduced international mobility of women in ICT sectors. Our findings follow a similar direction as stated above and confirm the main findings of the previous literature, also in the Spanish context. Mobility of women in ICT careers is deeply affected by personal reasons, especially family reasons. Female careers are subjected to male careers. Although most people interviewed, men and women, have said they have taken their partner into account in the decision of international mobility, previously, women had to negotiate much more than men with their partners and families. Women tend to shorten their time away in order to reconcile family and their professional career. To look for shorter periods away or to seek grants that enables the mobility of the whole family seem to be strategies more likely to be experienced by women than men. Most professional women have partners that also work in ICT sector, confirming that professional women are more likely to have partners in the same area of knowledge than men. Couples break up more easily due to woman’s mobility because men are less likely to cope with the changes that mobility involves. Thus changes are primarily related to social or family pressures and men’s difficulties to adapt to their new contexts. Finally, as clarified by the literature, women are more likely to delay motherhood than men fatherhood as they have fewer children than interviewed men. The roles traditionally associated with women also negatively influence the ability to take risky decisions. Insecure job situations will be rejected, such as mobility projects that can threaten employment situations, even when their employment is already precarious.

4. CONCLUSIONS The data highlighted that women are a minority in the international mobility of highly skilled workers. Even more so in the ICT sector. From our findings we can deduce that the minor international mobility of women in ICT areas of knowledge is not related to a reduced capacity for Technology. On the contrary, they suggest the existence of inequalities and the imposition of self-restrictions that relegate women to secondary roles in their professional careers.

240


The qualitative research carried out has allowed us to delve into the reasons for some of these inequalities in relation to mobility. Confirming most of the previously reviewed literature about mobility in Scientific Careers and gender in Spain, personal and family gendered factors, above all, determine international mobility of Women in ICT areas of knowledge. Women careers are more likely to be subject to male careers, they shorten their time away in order to reconcile their careers with family life, they are more likely to have partners in ICT than men, couples break up more easily with the mobility of the woman as men are less likely to cope with the mobility changes involved, and, finally, they delay childbearing and have fewer children than men.

REFERENCES Addis, E., 2004. Gender in the publication process: Evidence, explanations, and excellence, Gender and Excellence in the making, Directorate General for Research Information and Communication Unit. Ackers, L., 2004. Managing Work and Family Life in Peripatetic Careers: The Experiences of Mobile Women Scientists in the European Union, Centre for the Study of Law and Policy in Europe Research Report nº 1 Ackers, L., 2005. Gender, Mobility and Progression in Science Career: MOBISC Summary Report UK, Centre for the Study of Law in Policy in Europe Castaño, C. (Coord.), 2008. La Segunda Brecha Digital, Cátedra, Madrid Comisón Europea (2008). Women in ICT. Status and the way ahead, EC, Information Society and Media. Docquier et al., 2007. A Gendered Assessment of the Brain Drain, IZA Discussion Paper No. 3235 González Ramos A.M., 2009. La carrera profesional de las investigadoras jóvenes: Un camino lleno de posibilidades Revista CT+S, nº. 12, vol. 4: 31-54 OECD, 2007. International Migration Outlook: Annual Report 2007 Edition, OECD, Paris OECD, 2008a. The Global Competition for Talent. Mobility of the Highly Skilled, OECD OECD, 2008b. A profile of Immigrant Populations in the 21st Century. Data from OECD Countries, OECD Pérez Sedeño, E. 2003. La situación de las mujeres en el sistema educativo de ciencia y tecnología en España y su contexto internacional. Programa de análisis y estudios de acciones destinadas a la mejora de la calidad de la enseñanza superior y de actividades del profesorado universitario. MEC, Madrid. Stalford, H., 2005. Parenting, care and mobility in the EU. Issues facing migrant scientist, Innovation, vol. 18, nº 3, pp. 361- 380 Skachkova, P. 2007. Academia careers of immigrant women professors in the U.S. Higher Education, 53: 697-738 Teichler, U. and Maiworm, F., 1997. The ERASMUS Experience. Major findings of the ERASMUS Evaluation Research Project, CEC, Brussels This work has been realised due to funding from the Spanish Ministry of Education and Science (EA2008-0153)

241

ISBN: 978-972-8924-82-9 © 2009 IADIS

EMPIRICAL STUDY OF ROOT CAUSES OF CHANGE IN IT IMPLEMENTATION PROJECTS Ossa Elhadary Felician College

ABSTRACT In this paper the author studied 58 IT projects implemented during the period 2006 - 2007 and attempted to categorize the root causes of changes in those projects as well as identify the frequency of occurrence changes driven by each of those categories. The author also tried to discover relationships between the number of changes occurring and various project variables like project size, duration, etc. The research concluded that 22% of the projects implemented experienced at least one change and that the 2 most frequent causes of change were “Changed conditions” (“unbudgeted and unanticipated changes to baseline assumptions as they pertain to site conditions and weather) and “Programming Evolution” (change in/to the underlying basis of design as captured and identified in the baseline). This implies that more needs to done to identify risks that might derail the project and proactively plan responses to mitigate those risks. The fact that “ “Changed Conditions” seem to be a strong driver to change as shown by this research is understandable as in today’s complex IT projects and for a successful implementation many prerequisites and conditions need to be met and when they are not, change to projects become inevitable. In this research the author was also able to show that the longer the duration between the creation of the SOW and the project start (kickoff meeting), the more likely that changes will occur in the project which implies that the longer the delay in project start the more there is a need to reevaluate some of the assumptions made during the planning phase. KEYWORDS IT Implementation, Project Management, Change

1.

INTRODUCTION

Change is inevitable in many projects and it would be wrong to assume that all projects will proceed exactly as planned without any change occurring. Effectively managing projects and the dynamics of change is critical to success (Pardu, 1996). After studying the cumulative effect of change in construction projects, McCally (2006) concluded that owners and contractors must jointly address the cost and effects of changes as and when they occur and agree on a method whereby these costs and effects can be reasonably estimated. Many companies implement a change management process in their overall project management methodology in order to better manage project change. Pappas (2006) sites a survey in 2005 of 411 companies that indicated that 55% of those companies used a structural change management methodology up from 34% in 2003. Tichacek (2006) identified six root causes of change in projects: 1) Programming evolution which he described as change in/to the underlying basis of design as captured and identified in the baseline. He believed that this is a very common driver especially given the vast technological developments that occur during the often long planning to execution. 2) Market conditions which he described as the unanticipated, unbudgeted shortage of material, labor, or capital that bid-up pricing. 3) Oversight errors and omissions in the contract document 4) Failures in performance which he defined as the failure to provide complete or timely delivery of contractually committed events. 5) Changed conditions which he described as the “unbudgeted and unanticipated changes to baseline assumptions as they pertain to site conditions and weather.

242


6) Time related which he described as the deliberate and proactive decisions to “purchase time” in the project execution plan.

2. PROJECT MANAGEMENT METHODOLOGY In 2006-2007, the author worked as an IT project manager for a Fortune 1000 systems integrator in the United States and during that period, he managed tens of IT implementation projects for that company. The author identified all the changes that occurred in those projects and categorized them according to the Tichacek’s (2006) categories. The author hypothesized that the longer the period from the development of the statement of work (SOW) until the project start (project kickoff meeting), the more likely that a change will occur. The author also hypothesized that the larger the project (in terms of number of budgeted hours), the more likely that a change will occur. The author followed the company’s established project management methodology in managing the assigned project. The project’s statement of Work (SOW) is created by the sales team and in it the project’s cope, cost, as well as any other relevant information are documented. Once the customer signs that document, it is considered the legal contract that binds the company and the customer during this engagement. The project manager is then assigned and he/she discusses the project with the internal team and then schedules a kickoff meeting with the customer to signal the official launch of the project. During or even before the project implementation and once a change is identified, the project manager documents the change in a change request form identifying the effect on scope, cost, and schedule, and then submits it to the sales team who add the selling price. The customer then signs the change request form and then the change is implemented.

3. DATA ANALYSIS The author already had all the project documents for the projects managed in 2006 and 2007 and he extracted from these documents a number of variables that are relevant to this study. The following information was then entered into Excel: 1) Project ID 2) Project cost 3) Number of changes 4) Description of change 5) Duration from creations of SOW to Kickoff meeting 6) Duration from kickoff meeting to project completion 7) Duration from creation of SOW to project completion. The project population consisted of 58 projects of various sizes and scopes. Of those projects 13 had a documented change form created. Accordingly 22% of the projects incurred change. Overall there were 16 change forms. The author then used Pearson correlation coefficient to assess if there is any correlation between the chance of change occurring and some of the other project variables. The results of the analysis are shown in table 1.0. Table 1. Correlation between project variables and number of project changes

Variable Project Cost Duration from creations of SOW to Kickoff meeting Duration from kickoff meeting to project completion Duration from creation of SOW to project completion.

Pearson Correlation Coefficient (r) .252 .299

Square of Correlation (R) .063 .09

p-value 6.0197E-08 5.21956E-09

.553

.306

1.25224E-09

.613

.376

1.5633E-14

243

ISBN: 978-972-8924-82-9 © 2009 IADIS

According to the above table and because of the very low p-value, there seems to be a correlation between the 4 variables and the number of changes occurring in projects. But after careful analysis of the data, the author noticed that one large project seemed to influence greatly the correlation between the project size and the number of changes. The author accordingly removed this project from the population and recalculated the Pearson Correlation Coefficient and the p-value (see table 2. ). Table 2. Correlation between project variables and number of project changes after removing the outlier project

Variable Project Cost Duration from creations of SOW to Kickoff meeting Duration from kickoff meeting to project completion Duration from creation of SOW to project completion.

Pearson Correlation Coefficient (r) .148 .229

Square of Correlation (R) .022 .053

p-value 8.1473E-10 3.81706E-09

.532

.283

2.82237E-09

.59

.384

4.9622E-15

According to the above table, the strongest correlations between the numbers of changes seem to be the following variables: 1. The duration between the creation of the SOW and the project completion 2. The duration between the creation of the SOW and the project kickoff meeting

3.1 Categories of Change Out of the 58 projects implemented, 13 projects experienced a change and overall 14 changes were reported. The most frequent causes of change were “Changed conditions” and “Programming Evolution” with both occurring 5 times, followed by “Failure in Performance” which occurred 3 times, and then one occurrence of “Market Conditions” (see Fig 1.).

Changed conditions Failure in performance Market conditions Programming evolution

Figure 1. Root causes of change

244


3.2 Examples of Change: The following are examples of some of the changes that occurred in the project as well as the root causes that drove those changes: • Changed Conditions: The customer did not finish the migration of the Oracle Enterprise Manager before the site visit as planned, and accordingly the consultant could not proceed with the scope during the maintenance window and had to spend more time to finish the scope. • Programming Evolution: the customer requested that the consultant spend 40 additional hours working on issues that were not identified in the scope. • Failure in Performance: a second consultant had to be engaged to finish the scope because the first consultant had issues with the implementation. • Market Conditions: the SOW had to be changed to replace the primary consultant because he retired.

4. CONCLUSION AND RECOMMENDATIONS This research studied 58 IT projects implemented in 2006-2007 and successfully categorized the root causes of change using the six categories developed by Tichacek (2006). The author was able to show that 22% of the projects implemented experienced at least one change. The 2 most frequent causes of change were “Changed conditions” (“unbudgeted and unanticipated changes to baseline assumptions as they pertain to site conditions and weather) and “Programming Evolution” (change in/to the underlying basis of design as captured and identified in the baseline). This implies that more needs to done to identify risks that might derail the project and proactively plan responses to mitigate those risks. This research confirms Tichacek’s (2006) position that “Programming Evolution” is a very common driver to change although it also shows that “Changed Conditions” seem to be as frequent in driving project change. This though is no surprise especially that in today’s complex IT projects and for a successful implementation many prerequisites and conditions need to be met. In this research the author was also able to show that the longer the duration between the creation of the SOW and the project start (kickoff meeting), the more likely that changes will occur in the project. This was further emphasized by the fact there was also a significant correlation between the number of changes and the duration between the creation of the SOW and the project completion. These findings imply that the longer the delay in project start the more there is a need to reevaluate some of the assumptions made during the planning phase.

REFERENCES Pardu, Bill, May 1996. Managing change in a project environment. CMA, Vol 70, No. 4, pp. 6 Bob M McCally, 2006. The Cumulative Effect of Change. AACE International Transactions, pp. CS51 Lorna Pappas, Apr 2006. The speed of change. PM Network, Vol 20, No. 4, pp. 42 Robert L Tichacek, 2006. “Root Causes": The Six Reasons for CHANGE. AACE International Transactions, pp. PM71

245

Posters


WEB COMPETENCIES IN SME Klára Antlová Business Faculty, Technical University Liberec Studentská 1, 46117 Liberec, Czech Republic

Tomáš Gregorovič, Jindřich Tandler, Lubomír Popelínský1 Faculty of Informatics, Masaryk University Brno Botanická 68a, 60200 Brno, Czech Republic

ABSTRACT The main goal of this study was to find rules that describe a correspondence between web presentation, ICT competencies and company long-term growth. We bring evidence that a long-term growth can be predicted from the web presentation with high accuracy. We also show which attributes of the web page are the most important for that prediction. This investigation helps to SMEs to improve their competitiveness, better communication with customers and internet marketing strategy. KEYWORDS Web presentation, small and medium enterprises, ICT competencies, internet marketing, long term growth, data mining.

1. INTRODUCTION This article explores the competencies needed by the entrepreneurial small firm owner/manager for the effective use of the internet in their marketing activities. The research reviews a theoretical model of appropriate competencies and presents the findings of an empirical research in thirty organizations to establish the value of this model. This article goes on to propose a marketing strategy for developing web competencies. The thirty companies have been analyzed during qualitative research. The data has been collected through interviews with managers/owners. During the company existence the employees in SMEs developed competencies which are connected with using of ICT. A number of authors see the competence as the knowledge, skills and attitudes used by employees in the performance of work (Sanchez, 1998; Post, 1997). Such competencies require individuals or managers who posses capabilities in various areas. Hinckley and Perl (1996) interviewed managers and established an understanding of competencies required by managers. They indentified three areas of competencies: intra personal, interpersonal and organizational competence. As a part of their organizational competence is necessity to cooperate and communicate with employees, partners and customers. Information and communication technology and also using internet for the communication and marketing play important role as tool for these competencies. In the article (Antlova and Popelínský, 2008) the authors have searched what ICT competencies contribute to successful implementation of information systems and which competencies also help to increase their competitiveness. Next table explains shortly the examined ICT competencies (Peppard and Ward, 2004). The successful and growing enterprises developed especially competencies connected with business and information strategy (competencies number 1 and 2), competence connected with improving knowledge and skills of employees (competence number 6) and competence connected with electronic communication (number 7). The results from this survey indicated importance of sharing knowledge inside companies, business and information strategy and electronic communication with customers. 1

The authors have been supported by Faculty of Informatics, Masaryk University, Brno

249

ISBN: 978-972-8924-82-9 © 2009 IADIS

Table 1. ICT competencies. Number

Area of ICT competencies

1 2

Formulation of business strategy, the potential of new technologies, defining of ICT management Formulation of information strategy, the portfolio of investment, relevant R&D into how ICT can be used Information, application and technology architectures, organizations structures and processes to manage the resources Identifying and planning the benefits from ICT investment, monitoring, measuring and evaluating the benefits derived from ICT investment Implementing of information, systems and technology solutions that satisfy business needs, business continuity and security Recruiting, training and deploying appropriate staff and ensure technical, business and personal skills meet the needs of the organization Developing and maintaining the relationship with customers

3 4 5 6 7

Data mining aims in extraction of interesting – i.e. non-trivial, implicit, previously unknown and potentially useful - information or patterns from data (Han and Kamber, 2006) by means of statistics, vizualisation or machine learning (Mitchell, 1997; Witten and Frank, 2005). Data mining has been succesfuly used in market analysis and management, e.g. for target marketing, customer relation management, or market segmentation but also for and text and web analysis. The goal of web data mining is to discover useful knowledge from Web hyperlinks, page contents, and usage logs (Liu, 2008; Bichler, 2006). In this paper we apply web content mining and analysis of hyperlinks for finding dependency between long-term growth of a company, ICT competencies and its web presentation.

2. WEB COMPETENCIES In this work we focus especially on extracting information from web pages that can be obtained without human assistance (Liu, 2008). The only exception is attribute Look which contains evaluation of a web presentation set by human. Attributes extracted from web presentation and used in this work can be split into four groups. First two attributes - Look, Design - concern appearance. The second group contains features that characterize contents of a presentation. They concern a language of a web page and particular items/links that may be useful – work positions, an existence of authenticated area and a certificate, references from costumers and partners, news, information to download, a sitemap, e-order, web search and ICQ number. The third group describes the text – in menu or in description of the profile of the company. The last group characterizes importance of the web presentation in terms of web search engines Google and Yahoo. PageRank (from a link analysis algorithm used by the Google) assigns a numerical weighting to each element of a hyperlinked set of documents with the purpose of measuring its relative importance within the set. Backlinks is equal to the number of backlinks to the web page, obtained from Yahoo.

3. DATA AND METHODS From 30 companies explored in the previous work we removed two companies that have no web pages. We added information about ICT competencies and the attribute that classified a company as growing or not growing. Companies were uniformly distributed - 14 companies with long-term growth, 14 without longterm growth. We employed two classes of methods - classification/prediction and association mining. All experiments have been performed with Weka package (Witten and Frank 2005). For finding associations we used Apriori algorithm. For classification we employed several learning algorithms - Naïve Bayes, Support Vector Machines, the instance-based learner kNN(N=3) and decision tree learner J48. As evaluation criterion we have chosen overall accuracy, i.e. the ratio of correctly classified companies to the number of all companies. We used leave-one-out method for testing (27 instances used for learning, 1 for testing, for all permutations, an overall accuracy has been computed as an average of those 28 experiments). We also

250


checked various pre-processing techniques, namely for finding a small subset of important attributes, for ranking important words.

4. PREDICTION OF A LONG-TERM GROWTH FROM WEB PAGES To predict long-term grow from web pages we performed several experiments. In the first one we did not exploit text. In the second we used words that appeared in the menu on the web page. In the last experiments we added words and/or lemmata that were frequent in the whole collection of web pages. In the first experiment, when ignoring any text information, the decision tree learner J48 returned the best results and reached accuracy 82%. The decision tree contained only two attributes, Backlinks (a number of backlinks of yahoo) and Workpositions. In the next experiment we added words that appear in menu and ignored the rest of the text. Accuracy did not increased (the highest accuracy was 75%). The most significant attribute was again Baclinks and reference and profil were the most significant words. In the last experiments we used only text (words that appear on the web page) or text plus lemmatized text (a lemma of nouns, adjectives etc. is a nominative in singular, a lemma of a verb is its infinitive form). The best results has been reached with Naïve Bayes classifier and Support Vector Machines (SMO). The highest accuracy augmented to 82.1%. Besides Backlinks and Workpositions mentioned in 3.1.1, among the most relevant attributes for longterm growing companies the principal role plays web design. Successful companies mostly used cascade styles (attribute Design). They also frequently offer information about product to download and the site map. Strong positive correlation between PageRank and long-term growth has been observed - the higher PageRank, the higher probability for the long-term growth.

5. PREDICTION OF LONG-TERM GROWTH FROM WEB PAGES AND ICT COMPETENCIES In article (Antlová and Popelinský, 2008) a prediction of long-term growth from ICT competencies been described. We wanted to find whether information from web pages in combination with ICT competencies would result in an increase of prediction accuracy. However, it has not been observed. Actually, ICT competencies have much higher influence to prediction of long-term growth than any feature extracted from web pages. First we tried to predict ICT competencies from web pages. We took each ICT competence and explored how it depends on attributes of web presentation. Results of this survey indicate which web competencies are strongly connected with ICT competencies of SMEs. These competencies are: • competence number 2 covering business process design (how ICT can deliver best practice in operational processes and organizational activities), • competence number 5 - technology standards (it means developing and maintaining appropriate standards, methods, controls and procedures for the use of ICT and associate resources), • competence number 6 - ICT staff development (it means recruiting, training and deploying of appropriate staff and ensure technical, business and personal skills), • competence number 7 electronic communication with customers and suppliers. Because the SMEs are limited in their financial and human resources, the predicted value of successful web presentation is extremely useful. Results of this survey indicate the importance of business, information and knowledge strategy for their competitiveness and acknowledge the results of previous survey (Antlová and Popelinsky, 2008) where the ICT competencies supporting long term growth have been analyzed. In this experiment we looked for associations between web competencies, ICT competencies and longterm growth, we employed Apriori algorithm for mining association rules. An association rule consists of a conjunction of attribute-value couples (called body, and antecedent) and a head (a consequent, a couple attribute-value). Here we introduce three most interesting rules. For 9 companies, if the web pages were built using cascade styles, those companies are successful even though words international and modern are missing on those web pages.

251

ISBN: 978-972-8924-82-9 © 2009 IADIS

18 companies that do not use words global, major, top in their profile and do not have a site map, do not have investment into IT, i.e. ICT competence 1.3. 11 companies that do not use words world, system in their profile and do not have an item Services in the main menu, do not have ICT competence 6.2 – technological standards.

6. CONCLUSION We proved that a long-term growth can be predicted from the web presentation with accuracy higher than 80%. We also showed which attributes extracted from web presentation are the most important for that prediction. We also showed which information and communication technology competencies can be learned from the web presentation. As a result, this investigation can help SMEs to improve their competitiveness, communication with customers and internet marketing strategy. Some actual web pages of cooperation companies are now designed with regards to the results of this study. The findings of this study show that the company’s web presentation is not just commercial information on the internet but if it is well designed, it can improve communication with customers and marketing strategy and support long term growth. From web presentation it is also possible to predict whether the company puts strength on proper business strategy and is using the possibilities of ICT. These activities need employees with proper ICT competencies. Therefore the managers or owners of SMEs should support educational activities in this area and it is also the challenge for the universities how to help SMEs to increase competitiveness with the educational trainee. Also the importance of language mutations has been proved. This study did not involve the variety of industries (some company are service oriented and the others are producers). This different view on the companies connecting with type of industry and services can bring some other interesting results. This requires further research particularly to identify which competencies are relevant for service oriented or producing companies.

REFERENCES Antlová, K., Popelínský, L., 2008, Predicting information system competencies in small and medium enterprises with data mining methods, IASK-2008,. Madrid, Spain, pp.133-139 Bichler, M., Lin, K.L., 2006. Service – Oriented Computing. Computer, Vol.39, No.3, pp.99-101. Caldeira, M.M., Ward, J.M., 2001. Using Resource-Based Theory to Interpret the Successful Adoption and Use of ICT in SME. The 9th European Conference on IS, Bled, Slovenia, June 27-29, 2001. Han J., Kamber M., 2006. Data Mining: Concepts and Techniques. Elsevier 2006. Hinckley, S., Perl, V., 1996. The Competence of Global manager. Global Management, Vol. 1, pp.138-148. Kamvar, S.D. 2003. Extrapolation Methods for Accelerating PageRank Computation, Proc. 12th Int. World Wide Web Conf., ACM Press, Liu B., 2008. Web Data Mining. Springer, Verlag 2008. Levy M., Powell P., 2005. Strategies for Growth in SMEs, Butterworth Heinemann, Oxford, 2005. Mitchell T., 1997. Machine Learning. McGraw Hill, New York, 1997. Post, H.A., 1997. Building a Strategy on Competencies, Long range Planning, Vol.30, No.5, pp. 733-740. Peppard, J., Ward, J., 2004. Beyond Strategic Information Systems: Towards an IS Capability, The Journal of Strategic Information Systems, Vol. 13, 167-194, 2004. Sanchez, D., 1998. Global Selling with Local Flavour, Sales and Marketing Management, Vol. 150, No. 8,pp. 26-28. Spiliopoulou, M., 2000. Web Usage Mining for Site Evaluation, Comm. ACM, vol. 43, no.8, pp.127-134. Witten, I.H., Frank, E., 2005. Data Mining. Practical Machine Learning Tools and Techniques. Elsevier 2005. http://www.cs.waikato.ac.nz ml/weka

252

AUTHOR INDEX Ahmed, A. ......................................................147 Anderson,K. ..................................................198 Antlová, K. ....................................................249 Bagnara, S. ....................................................221 Blackburn, T.....................................................20 Blignaut, S........................................................67 Bonadia, G. ....................................................184 Borglund, E. ...................................................198 Broin, U............................................................75 Buckner, K. ....................................................189 Caldwell, B.....................................................229 Danielson, M. ...................................................58 Ekenberg, L. .....................................................58 Elhadary, O.....................................................242 Els, C. ...............................................................67 Fielden, K. .....................................................105 González, A. ..................................................237 Gregorovič, T. ...............................................249 Gripenberg, P. ................................................193 Haftor, D................................................... 35, 203 Handri, S.....................................................83, 91 Heikkinen, M..................................................155 Hrycej, N. ......................................................189 Jouhki, J. ........................................................121 Kehal, M.........................................................213 Kitchens, F. ....................................................113 Kivunike, F.......................................................58 Kudo, Y. ...........................................................91 Lacroux, F. ....................................................203 Leahy, D. ..........................................................75 Malcolm, P. ....................................................105 Mat-jizat, J. ....................................................175 McKay, E. ......................................................175 Menezes, E. ...................................................184 Nakamura, K. ...................................................83 Necessary, J. ..................................................113 Nodin, N. ........................................................233 Nomura, S...................................................83, 91 Ogushi, C........................................................184 Palmen, R. ......................................................237 Paskaleva, K. ..................................................139 Pethe, A. .........................................................129 Popelínský, L. ................................................249 Pozzi, S...........................................................221 Rothwell, A. .....................................................51

Sandberg, K. .................................................. 198 Sirkemaa, S. ........................................... 198, 225 Sjøvoll, J. ......................................................... 28 Swatman, P. ...................................... 20, 98, 129 Tan, K. .............................................. 20, 98, 129 Tan, M................................................................ 3 Tandler, J. ...................................................... 249 Tiihonen, T....................................................... 11 Touru, A. ......................................................... 11 Tusubira, F. ...................................................... 58 Vartiainen, T. ................................................... 43 Vergés, N. ...................................................... 237 Vidal, P. ........................................................ 203 Wahlberg, O. ................................................. 198 Walsh, C......................................................... 163 Wohn, D......................................................... 209 Zimic, S.......................................................... 180

SECTION II


WEB BASED COMMUNITIES 2009


ii

TABLE OF CONTENTS FOREWORD

v

PROGRAM COMMITTEE

vii

KEYNOTE LECTURES

xi

SPECIAL TALK

xii

FULL PAPERS A PROCESS TO MANAGE CORPORATE KNOWLEDGE USING SOCIAL NETWORKS: A CASE STUDY Ricardo Araújo Costa, Rafael de Albuquerque Ribeiro, Silvio Romero de Lemos Meira and Edeilson Milhomem da Silva “BECAUSE I KNOW YOU”: ELEMENTS OF TRUST AMONG PSEUDONYMOUS BLOGGERS

3

11

Vanessa P. Dennen

FOSTERING STUDENTS’ LEARNING OF COURSE MANAGEMENT SYSTEMS BY USING BLENDED LEARNING

19

Ahmed El-Gamal and Thomas Köhler

SOCIAL LEARNING FOR UNIVERSITY STAFF

27

Francis Brouns, Adriana J. Berlanga, Sibren Fetter, Marlies E. Bitter-Rijpkema, Jan M. van Bruggen and Peter B. Sloep

PULLING DOWN THE WALLED GARDEN: TOWARDS A PARADIGM FOR DECENTRALIZED SOCIAL NETWORK MANAGEMENT

35

Stefano Bortoli, Themis Palpanas and Paolo Bouquet

SHARING EXPERIENCES USING WEB 2.0 AND KNOWLEDGE PORTALS FOR ADVANCING WEB COMMUNITIES

43

Nik Bessis, Mitul Shukla and Zhihua Lai

THE USE OF AN ONLINE HEALTH COMMUNITY ON OVERWEIGHT: A MEMBER PERSPECTIVE

51

Åsa Smedberg 59

THE NEED FOR TWO COMMUNITIES John S. Murnane

A SOCIAL CAPITAL PERSPECTIVE OF TRAVELERS’ PARTICIPATION IN WEB-BASED SOCIAL NETWORKING COMMUNITIES

67

Linda S L Lai

KNOWLEDGE MANAGEMENT AND SHARING IN OPEN SOURCE COMMUNITIES S. L. Toral, F. Barrero and M. R. Martínez-Torres

iii

75

SHORT PAPERS A MEASUREMENT MODEL ON BLOGGERS’ SELF-DISCLOSURE

87

Jih-Hsin Tang and Cheng-Chung Chen

EXPERIENCING CHANGE THROUGH THE ADOPTION OF NEW DIGITAL TECHNOLOGY

91

Mitul Shukla, Nik Bessis and Alfredo Gaitan

ANALYSIS OF ONLINE PARTICIPATION IN OPEN SOURCE COMMUNITIES

96

M. R. Martínez-Torres, S. L. Toral and F. Barrero

SEARCHING FOR MARKET TRENDS AND BUSINESS OPPORTUNITIES IN WEB COMMUNITIES

102

Graziella Martins Caputo and Nelson Francisco Favilla Ebecken

KNOWLEDGE SHARING THROUGH ONLINE COMMUNITIES OF PRACTICE: THE CASE OF LINUX PORTS TO EMBEDDED PROCESSORS

107

S. L. Toral, F. Barrero and M. R. Martínez-Torres

CYCLIC ENTROPY OPTIMIZATION OF A SCALE-FREE SOCIAL NETWORK USING EVOLUTIONARY ALGORITHM

113

Maytham Safar, Nosayba El-Sayed and Khaled Mahdi

DEVELOPMENT OF A CUSTOMER SATISFACTION MODEL FOR ENABLING E-COLLABORATION IN SECOND LIFE

119

Mitul Shukla, Nik Bessis, Marc Conrad and Gordon Clapworthy

CHARACTERIZING COLLABORATIVE SOCIAL NETWORKS USING CYCLIC ENTROPY, CASE STUDY: WIKIPEDIATM

125

Khaled Mahdi, Lama Jammal and Maytham Safar

HOW DOES SERVICE QUALITY AFFECT CUSTOMER SATISFACTION IN PERSONALIZED COMMUNITY SERVICE USAGE?

131

Su Jin Kim and Bong Gyou Lee

USING A SOCIAL NETWORK TO SUPPORT PROJECT MANAGEMENT

136

Paulyne Jucá, Ricardo Costa and Silvio Meira

A MULTIDIMENSIONAL MODEL OF TRUST AND ITS ANTECEDENTS FOR KNOWLEDGE SHARING IN VIRTUAL COMMUNITIES

140

Abel Usoro and Grzegorz Majewski

REFLECTION PAPERS RIMI ONLINE COLLABORATIVE COMMUNITY

147

Amy S. C. Leh

CLASSIFYING KNOWLEDGE ON SOCIAL NETWORKS: A REAL CHALLENGE Ricardo Costa, Paulyne Jucá and Silvio Meira, Isabel Mendonça and Haidee Lima

AUTHOR INDEX iv

151

FOREWORD These proceedings contain the papers of the IADIS International Conference Web Based Communities 2009, which was organised by the International Association for Development of the Information Society in Algarve, Portugal, 21 – 23 June, 2009. This conference is part of the Multi Conference on Computer Science and Information Systems 2009, 17 - 23 June 2009, which had a total of 1131 submissions. The mission of this conference is to publish and integrate scientific results and act catalytically to the fast developing culture of web communities. The conference invites original papers, review papers, technical reports and case studies on WWW in particular the emerging role of so-called WWW-Based Communities. The following five main topical areas made of twenty-seven topics have been object of paper and poster submissions: A. The history, architecture and future of virtual communities 1. From mobility to connectivity 2. Identity and augmented ideologies 3. Visionary web architectures, implanted computers 4. Network revolutions, post-colonial and post-modern societies 5. Escaping from reality, virtual reality and multi-user games 6. Towards alternative ways of presence B. Group processes and self-organization 1. Tele-democracy, morality, netiquette 2. Social networks, tribal- and open communities, peace education 3. Computer mediated-, hyper- and narrative communication, woven stories 4. MUDs, MOOs and avatars 5. Hosting web-based communities 6. Nationalities, ethnicities and gender effects C. Cyborgs, teleworking, telemedicine, art games and learning communities 1. Fading hierarchies and epistemic dictatorship 2. Distributed cognition, the electronic cortex and constructivism 3. Community directories 4. Mechanic world, organic computer 5. Agents and the vectorized self 6. Beyond metaphors: imagining and representation 7. Communizing as a marketing approach D. Expanding markets through virtual communities 1. The WWW as digital market place 2. The enterprise as a learning community 3. The learning as a road map for business 4. Universities as online communities 5. Business-to-business communication in profit- and non profit sectors

v

E. Virtual communities for people with special needs 1. Access to public spaces 2. Accessibility and long-term disabilities 3. Virtual communities in health care The IADIS Web Based Communities 2009 Conference had 53 submissions from more than 22 countries. Each submission has been anonymously reviewed by an average of four independent reviewers, to ensure the final high standard of the accepted submissions. The final result was the approval of 10 full papers, which means that the acceptance rate was below 19 %. A few more papers have been accepted as short papers, reflection papers and posters. Best papers will be selected for publishing as extended versions in the "International Journal of Web Based Communities" (IJWBC); ISSN: 1477 – 8394, and also in other selected journals, including other journals from Inderscience. The conference, besides the presentation of full papers, short papers, reflection papers and posters, also included two keynote presentations from internationally distinguished researchers. We would therefore like to express our gratitude to Dr. Bob Appelman, Indiana University, USA and Laurel Papworth, Social Network Strategist, Australia, for accepting our invitation as the conference keynote speakers. Also thanks to the special talk given by Nian-Shing Chen, National Sun Yat-sen University, Taiwan. As we all know, organising a conference requires the effort of many individuals. We would like to thank all members of the Program Committee, for their hard work in reviewing and selecting the papers that appear in the proceedings. This volume has taken shape as a result of the contributions from a number of individuals. We are grateful to all authors who have submitted their papers to enrich the conference proceedings. We wish to thank all members of our local organizing committee, sponsors, delegates, invitees and guests whose contribution and involvement are crucial for the success of the conference. Last but not the least, we hope that everybody will have a good time in Algarve, and we invite all participants for the next year edition of the IADIS International Conference Web Based Communities 2010, that will be held in Freiburg, Germany. . Piet Kommers University of Twente, The Netherlands Web Based Communities 2009 Conference Program Chair Piet Kommers, University of Twente, The Netherlands Pedro Isaías, Universidade Aberta (Portuguese Open University), Portugal Nian-Shing Chen, National Sun Yat-sen University, Taiwan MCCSIS 2009 General Conference Co-Chairs Algarve, Portugal June 2009

vi

PROGRAM COMMITTEE WEB BASED COMMUNITIES CONFERENCE PROGRAM CHAIR Piet Kommers, University of Twente, The Netherlands

MCCSIS GENERAL CONFERENCE CO-CHAIRS Piet Kommers, University of Twente, The Netherlands Pedro Isaías, Universidade Aberta (Portuguese Open University), Portugal Nian-Shing Chen, National Sun Yat-sen University, Taiwan

WEB BASED COMMUNITIES CONFERENCE COMMITTEE MEMBERS Adriana Berlanga, Open University of The Netherlands, Netherlands Alan Smith, University of Southern Queensland, Australia Alexander Stocker, Graz University of Technology, Austria Alison Varey, Napier University, UK Allan Yuen, The University of Hong Kong, Hong Kong Annabell Preussler, Universität Duisburg-Essen, Germany Apostolos Gkamas, University of Patras, Greece Barbara Kump, Graz University of Technology, Knowledge Management Institute, Austria Carlo Torniai, Simon Fraser University, Canada Chai Ching Sing, National Institute of Education, Singapore Christin Seifert, Know Center, Austria Christos Bouras, University of Patras, Greece Christos Georgiadis, University of Macedonia, Greece Claudia Thurner, Know-Center, Austria Daniel Tan, Nanyang Technological University, Singapore Deniz Deryakulu, Ankara University, Turkey Dragan Gasevic, Athabasca University, Canada Elias Pimenidis, University of East London, UK

vii

Eliza Stefanova, St. Kl. Ohridski University of Sofia, Bulgaria Enyedi Szilard, Technical University of Cluj-Napoca, Romania Eugenia Kovatcheva, University of Sofia, Bulgaria Eva Heinrich, Massey University, New Zealand Fatma Cemile Serce, Atilim University, Turkey Ferda Nur Alpaslan, Middle East Technical University, Turkey Flavio Soares Correa da Silva, University of Sao Paulo, Brazil Francis Brouns, Open University, Netherlands Frederico Figueiredo, Nokia Siemens Networks, Portugal Gal Springman, Language Connections Organization, Israel Gayle Davidson Shivers, University of South Alabama, USA George Dafoulas, Middlesex University, United Kingdom Giannis Koutsonikos, Technological Educational Institute of Patras, Greece Gisela de Clunie, Universidad Tecnológica de Panamá, Panamá Gisela Granitzer, Know Center Graz, Austria Giuliana Dettori, Istituto per le Tecnologie Didattiche, Italy Grigorios Beligiannis, University of Ioannina, Greece Honoriu Valean, Technical University of Cluj-Napoca, Romania Huang-Yao Hong, National Chengchi University, Taiwan Ilias Karasavvidis, University of Thessaly, Greece Ines Buchsteiner, Styria Medien AG , Austria J. Michael Spector, Florida State University/University of Georgia, USA Jaakko Kurhila, University of Helsinki, Finland Jan Frick, Stavanger University, Norway Jan Marco Leimeister, Kassel University, Germany Jay Wilson, University of Saskatchewan, Canada John Murnane, The University of Melbourne, Australia John Potter, Institute of Education, London UNiversity , UK Jon Dron, Athabasca University, Canada Jose Jesus Garcia Rueda, Universidad Carlos III de Madrid, Spain Jose Luis Sierra Rodriguez, Universidad Complutense de Madrid, Spain Kalogiannakis Michail, Technological Educational Institute of Crete, Greece Kevin Burden, The University of Hull, UK

viii

Kirsti Lindh, University of Tampere, Finland Konrad Morgan, School of Applied Media and Information Technology, NAIT, Canada Konstantinos Giotopoulos, University of Patras, Greece Krassen Stefanov, St. Kl. Ohridski University of Sofia, Bulgaria Lawrie Hunter, Kochi University of Technology, Japan Lee Chien-Sing, Multimedia University, Malaysia Lee Fong-Lok, The Chinese University of Hong Kong, Hong Kong Ljuan Marko Gashi, University of Novi Sad, Serbia Lorna Uden, Staffordshire University, United Kingdom Marc Spaniol, Max-Planck-Institut für Informatik, Germany Marco Kalz, Open University of the Netherlands, The Netherlands Marcus Specht, Open University of The Netherlands, Netherlands Maria Grazia Ierardi, CNR, Italy Martin Gonzalez, Tecnológico de Monterrey, Mexico Martin Llamas-Nistal, ETSI Telecomunicación, Spain Michael Granitzer - Know Center Graz, Austria Michael Kerres, University Duisburg-Essen, Germany Michalis Xenos, Hellenic Open University, Greece Nikolina Nikolova, St. Kl. Ohridski University of Sofia, Bulgaria Nikos Mastorakis, University of Sofia, Bulgaria Nina Kahnwald, Technische Universität Dresden, Germany Nopphol Pauswasdi, Mahidol University, Thailand Oliver Bohl, Kassel University, Germany Pablo Moreno-Ger, Universidad Complutense de Madrid, Spain Panayiotis Fouliras, University of Macedonia, Greece Paola Forcheri, IMATI-CNR, Italy Patrick Hoefler, Know Center, Austria Peter Albion, University of Southern Queensland, Australia Peter Kraker, Know Center, Austria Peter Mikulecky, University of Hradec Kralove, Czech Republic Peter Sloep, Open University of The Netherlands, The Netherlands Ralf Klamma, RWTH Aachen University, Germany Raúl V. Ramírez Velarde, Tecnologico de Monterrey, Mexico

ix

Robert Brazile, University of North Texas, USA Sandra Lovrencic, University of Zagreb, Croatia Shakuntala Banaji, London Knowledge Lab, UK Sibren Fetter, Open Universiteit Nederland, The Netherlands Slavi Stoyanov, Open University of The Netherlands, The Netherlands Sobah Abbas Petersen, Norwegian University of Science and Technology, Norway Stefania Bandini, University of Milano-Bicocca, Italy Stefanie Lindstaedt, Know Center Graz, Austria Stylianos Hatzipanagos, King's College London, UK Tatyana Pashnyak, Bainbridge College, USA Thomas Köhler, Universität Potsdam, Germany Tiberiu Letia, Technical University of Cluj-Napoca, Romania Tobias Ley, Graz University of Technology, Austria Ulrich Thiel, Fraunhofer IPSI, Germany Vaggelis Kapoulas, Research Academic Computer Technology Institute, Greece Vanessa Dennen, Florida State University, USA Vassilis Kollias, University of Thessaly, Greece Vassilis Poulopoulos, Research Academic Computer Technology Institute, Greece Vicente Luque Centeno, Universidad Carlos III de Madrid, Spain Victor Lopez Cabrera, Technological University of Panama, Panama Viktoria Pammer, Graz University of Technology, Austria Violeta Damjanovic, Salzburg Research, Austria Violeta Vidacek-Hains, University of Zagreb, Croatia Vladimir Bures, University of Hradec Kralove, Czech Republic Volker Patent, Open University, UK Wang Qiyun, National Institute of Education, Singapore Wilhelmina C. Savenye, Arizona State University, USA Witold Abramowicz, Poznan University of Economics, Poland Wolfgang Kienreich, Know Center Graz, Austria

x

KEYNOTE LECTURES COMMUNITIES IN THE LABYRINTH Dr. Bob Appelman Indiana University USA

ABSTRACT Dr. Appelman, will discuss the World Wide Web as a labyrinth of paths to information, communities, and personal exploration. His talk is intended to be not only the unveiling of a tapestry of strategies, theories, and practice, but also a whimsical journey through what some people call the new reality. This keynote is intended to be inclusive, challenging, informative, and a discussion that can set the tone for this conference

THE KEY FEATURES OF SOCIAL MEDIA SITES THAT BUILDS ONLINE COMMUNITY Laurel Papworth Social Network Strategist Australia

ABSTRACT

Simply uploading forum software is not enough to create community. This session focusses on how to build sociability across platforms and the challenges and opportunities to improve behaviour across business and government social networks online.

xi

SPECIAL TALK A CYBER REAL CYBER MODEL FOR RUNNING INTERNATIONAL CONFERENCES Professor Nian-Shing Chen National Sun Yat-sen University Taiwan

ABSTRACT

One of the main purposes for participants attending international conferences is to meet people, who are doing research in the same field/topic, for exchanging ideas, sharing experiences and even looking for joint collaborative research. However, the time is always very limited during the conference; many participants had an experience that once they found someone who shares the same research interests, it is time to say goodbye. Though, many people would say let’s keep in touch and correspond by emails for further exchanges and discussions, we all know it just does not work in this way for most of cases. How do we cope with this issue by taking the advantage of cyber community? I will introduce in this talk a cyber-real-cyber model for running an international conference in blended mode to solve this problem. The basic idea is to extend the time span of an international conference to both before the physical conference and after the physical conference. A web-based community conference platform with many Web 2.0 features has been developed to facilitate participants to start with communications and discussions with those authors they are interested before the physical conference. This would then shorten the social distance among these people who have been getting to know each other in cyber space, so when they finally meet together at the conference physically, they can have a better opportunity for arranging small group meetings and exchanges. Furthermore, after the conference, those people who have already established connections during the conference can continue to exchange and collaborate together via the asynchronous and synchronous cyber office environments supported by the web-based community conference platform.

xii

Full Papers

IADIS International Conference on Web Based Communities 2009

A PROCESS TO MANAGE CORPORATE KNOWLEDGE USING SOCIAL NETWORKS: A CASE STUDY Ricardo Araújo Costa, Rafael de Albuquerque Ribeiro, Silvio Romero de Lemos Meira Center for Informatics - Federal University of Pernambuco (UFPE) and Recife Center for Advanced Studies and Systems (CESAR)

Edeilson Milhomem da Silva Center for Informatics - Federal University of Pernambuco (UFPE)

ABSTRACT This case study describes the effects of using a Web Based Social Network (WBSN) approach to Knowledge Management in a Brazilian software development organization. During this work it was proposed a Knowledge Management process, with some metrics related to its phases. These metrics have been monitored since January 2008 and have been analyzed in order to verify the efficiency of this approach. In order to give a better understanding of the concepts related to Social Networks and Knowledge Management, it is presented a brief introduction to each one of them, including an evaluation of existing Knowledge Management approaches. KEYWORDS Enterprise Social Network, Knowledge Management, Process, Metrics.

1. INTRODUCTION One of the many ways to achieve industrial competitiveness is to manage and share efficiently knowledge built inside an organization. In this context, social networks have shown signs of being an efficient tool to proliferate individual and explicit knowledge, even improving the tacit knowledge dissemination, helping to capture organizational knowledge based on the knowledge of each of its employees. Staab (2005) has affirmed that social networks are a very good mechanism to promote more interactivity between individuals. The capturing of tacit knowledge trough interaction tools inside the Web Based Social Networks also allows extending this tacit knowledge. According to Davenport (1998), in order to create new knowledge is necessary to expose humans to new information, so they can process it and generate new knowledge inside their minds. This work presents the a.m.i.g.o.s, a Web Based Social Network (WBSN) environment, and the experience of its use as the main tool to foster the communication, collaboration and knowledge management inside C.E.S.A.R, an Innovation Institute located in Brazil. Besides this introduction, the reminder of this paper is organized as follows: a short explanation about knowledge management concepts, including its proposed process and metrics; definitions and benefits regard the use o Social Networks; the case study developed during this work, with a mapping between a Knowledge Management (KM) process and the a.m.i.g.o.s functionalities and effects of its use in C.E.S.A.R during the second phase of this case study; Finally it presents the concluding remarks and directions for future works.

2. KNOWLEDGE MANAGEMENT According to Choi and Lee (2003), Knowledge Management (KM) in a software corporation is an opportunity to create a default perception among software developers so they can interact, deal and share knowledge and experiences. The reduction in loss of Intellectual Capital from employees who leave the

3

ISBN: 978-972-8924-82-9 © 2009 IADIS

company; the cost reduction on the development of new products; and the increased productivity by making knowledge easily accessible to all employees are some of several benefits in using a Knowledge Management strategy. According to Wenger (Wenger, 2004), the Knowledge Management field has come to realize the importance of communities of practice as the social knowledge fabric. He argues that the ability to truly manage knowledge assets resides on actively involving the practitioners on the knowledge management process. Wenger (Wenger, 2004) identified three main characteristics in communities of practice. Domain provides a common focus and brings the community together; Community builds relationships that enable collective learning; and Practice anchors the learning in what people do, it represents the body of knowledge the members share and develop together. He affirms that cultivating communities of practice requires paying attention to all three elements.

2.1 Knowledge Management Metrics If in one hand researches show that there is already a good knowledge base on performance indicators for knowledge management (Bose, 2004), on the other hand Liebowitz (2000) reviewed some of the performance indicators present in the literature and stated that “many of the cited metrics lack "creativity" in terms of determining the size and growth of the organization's knowledge base”. Nonaka (1995) states that some managers have difficulties in fitting into the model of a knowledge driven company since they believe that the only useful knowledge is the quantifiable one and that the company is a sort of information processing machine. Ahmed (1999) also points the same difficulties for having higher level performance indicators, affirming that traditional performance measures have focused on outputs, whereas there is a need to look towards the enablers that lead to the production of results. This only reinforces the lack of sensitiveness a financial indicator suffers, since such categories of indicators are only able to provide the whole picture of the Knowledge Management area, being unable to measure every component area. In order to have a sound management of the whole Knowledge Management process, corporations need to be able to measure every stage of the whole process. Concerned with these limitations, apart from reviewing current indicators present on literature, Liebowitz (2000) proposed the following indicators: • The number of new colleague to colleague relationships spawned; • The reuse rate of “frequently accessed/reused” content; • The capture of key expertise in an online way; • The dissemination of knowledge sharing to appropriate individuals; • The number of knowledge sharing proficiencies gained; • The number of new ideas generating innovative products or services; • The number of lessons learned and best practices applied to create value-added; • The number of (patents + trademarks + articles + books + talks at conferences) / employees; • The number of “apprentices” that one mentors, and the success of these apprentices as they mature in the organization; • Interactions with academicians, consultants and advisors.

3. SOCIAL NETWORKS The Social Networks theory approaches the social relationships as nodes and links. Each node represents an actor within the social network, and each link represents a social connection between actors. There are many different ways to link these actors, each one related to the nature of the represented social network (Iacobucci, 1994). One of the reasons justifying the interest of organizations in social networks is how these networks are quite efficient to share the knowledge of each individual (Staab et al., 2005). Once that knowledge, that is relevant to the members of the Social Network, is documented, it can be reused, avoiding wasting of effort from employees.

4


According to Domingos (2001), the way that users publish information in social networks is impressive and without any precedent. Because using social networks is an efficient way to share and distribute individual knowledge, its use to support a knowledge management initiative has become a frequent approach. Moreover, using social networks can bring other benefits, like providing an interactive and informal environment where users can express themselves, easily enriching the organization memory.

4. A SOCIAL NETWORK AS A KNOWLEDGE MANAGEMENT TOOL In order to achieve a better result in its knowledge management strategy, C.E.S.A.R. has developed and incorporated the use of a WBSN tool named a.m.i.g.o.s, which was first deployed in October 2006. This new initiative is trying to add some human-oriented aspects to a system-oriented approach (Costa et al., 2008) Portuguese acronym for Multimedia Environment for Integration of Groups and Social Organizations, the main goal of a.m.i.g.o.s is to provide a software infra-structure to support the creation of WBSN. The a.m.i.g.o.s also intends to stimulate the knowledge creation and sharing by its members, providing many features in order to be used as a knowledge sharing tool (Costa et al., 2008).

4.1 Mapping a cyclic KM Process on A.M.I.G.O.S As one of the main goals of A.M.I.G.O.S. environment is to provide a tool to foster the knowledge acquisition, combination and sharing through the organization, it was made a comparison between the KM process proposed by Bose (2004) and the existing functionalities, generating the a.m.i.g.o.s KM process (Fig. 1), presented as follows.

Figure 1. a.m.i.g.o.s KM process

This proposed process is composed by 4 phases, each focusing in a set of functionalities and its usage. On the following sections the phases of Creation and Capture, Refinement and Storage, Management, and Dissemination are presented in more details.

4.1.1 A.M.I.G.O.S. Knowledge Creation and Capture Bose proposed that the knowledge comes from experiences and skills of the employees and must be stored in raw form to be useful to others members of the organization. In the a.m.i.g.o.s environment there are basically two ways to users create and capture knowledge. The first source of knowledge is the reported user’s experiences. Every user is encouraged to describe his experiences, lessons learned and any other information as stories, with its related stories, images, videos,

5

ISBN: 978-972-8924-82-9 © 2009 IADIS

audios or any other electronic document. These stories can, additionally, be related to specific communities, increasing the contextual information of the described experience.

4.1.2 A.M.I.G.O.S. Knowledge Refinement and Storage The refinement and storage of knowledge are made through a diversity of activities that take place inside the a.m.i.g.o.s. environment. Users are encouraged to develop new knowledge through dialogues about any subject related to a community. The entire dialogue is immediately stored and made accessible by any other member of the community, and, depending on permissions policy, to all a.m.i.g.o.s. users. The refinement can also occur inside stories created by users, where others can add more information, or even refute what is written. This also is immediately stored in the repository.

4.1.3 A.M.I.G.O.S. Knowledge Management Because it is basically a social network, there is not a tool in the a.m.i.g.o.s. environment that can be used to allow knowledge management made by a group of specific users, usually the KM team. Instead, several mechanisms were developed to allow the users to manage knowledge implicitly. The first mechanism is the rating tool. Every time a user reads a piece of knowledge (stories, topics, documents, objects, etc.), he can rate it so the users can be directed to the most relevant pieces of knowledge. The second mechanism is the stories sorting mechanism. Through these filters it is easy for users to identify which are the newest stories, the most recent commented, the most active (a mixture of most commented and most recent), and so on. Another efficient mechanism used to manage the knowledge is the folksonomy, which allows any user to classify any existing piece of knowledge with some special keywords, called tags. These keywords are used by users to filter and find knowledge related to a specific subject.

4.1.4 A.M.I.G.O.S. Knowledge Distribution According to Bose, an effective knowledge management needs that every piece of knowledge is accessible to every member in an organization, at anytime and in an organized way. In a.m.i.g.o.s., the distribution process is focused in searches for knowledge (stories, forum, topics, messages, communities and experts) according to the permissions set for each one of them. But what makes a.m.i.g.o.s. a different social network environment is the recommendation engine that runs in background according to actions taken by users. The base to this recommendation process is the profile inference module. Every content written or added (as electronic documents) to the a.m.i.g.o.s social network is analyzed and the relevant keywords, with its weight, are added to the user profile. Additionally, every story which was positively rated is also added to the user profile. So every user in the environment has two groups of relevant keywords, the keywords related to subjects which the user likes to write about, and another set of keywords related to subjects which the user is currently interested in reading. The mechanism of automation of the recommendations passed by the following developments: (1) recommendations for similar content based on the information that users usually post / write in communities, stories and objects (documents or sites), (2) recommendations for taking similar stories as a basis the stories read by the user, and (3) recommendations of relevant users (experts). These mechanisms aim to attend the Bose objectives, which are: foster the organization knowledge through the creation of new tacit knowledge (through exposing the employees to new explicit knowledge) and increase the employee qualification.

4.2 Results In July 2008 the second phase of the a.m.i.g.o.s case study was initiated. Because the previous phase had already indicated that the WBSN approach to KM was the better approach to C.E.S.A.R (Costa et al., 2008), the second phase main goal was to have a better understanding about the behavior of a.m.i.g.o.s’ users on the Knowledge Management perspective. To achieve this better understanding it was necessary to define some new metrics based on the work made by Liebowitz, in 2000. So, during this phase the following metrics were defined and monitored:

6


•

•

• • • • • • • • •

Number of new Knowledge Published: Which on the wiki approach is basically the number of file uploads and page creations, and on the a.m.i.g.o.s approach is the number o stories related and uploaded objects. Because it is needed a deeper evaluation to verify which stories or objects are new knowledge, this metric worked with every possible new knowledge. Number of Discussions: Which for the first initiative is the number of messages sent to the general mailing lists plus the number of messages sent to the engineering mailing lists, and for the a.m.i.g.o.s approach is the number of topics and messages initiated and posted on public communities. Number of Recommended Knowledge: Which is basically the number of recommendations on a.m.i.g.o.s but do not have an equivalent on the wiki and mailing list approach. Number of added contacts: This indicator is based on “the number of new colleague to colleague relationships spawned”; Number of times a topic or history is read: This indicator is based on “the reuse rate of ‘frequently accessed/reused’ content”. For this to be calculated it was defined that a story or topic is read if it stays open on the browser for at least five seconds; Number of stories and topics positively qualified: This indicator is based on “the capture of key expertise in an online way”. For this to be calculated, it is considered any story or topic with an average rate greater than or equal to three stars (which can vary from zero to five stars); Number of knowledge classifications: This indicator tries to measure how frequently the created knowledge is classified using tags. For this to be calculated, it was counted the number of tags applied to any item (stories, topics, objects and comments) inside the environment; Number of qualifications: This indicator tries to measure how often a knowledge item (stories and topics) is rated by users, independent of the rate; Number of access to elements through tags: This indicator tries to measure how often users look for items related to a specific subject, represented by a tag; Tagcloud visualizations: This indicator tries to measure how often users look for knowledge through the list of subjects available, which is represented by the environment tagcloud; Number of unique users: This indicator tries to measure what is the real impact of the system on the organization. For this it calculates how many different users log in and uses the system in a month.

Figure 2. a.m.i.g.o.s monthly metrics

When analyzing the knowledge creation metrics (Figure 2) it is possible to verify that there is a pattern on knowledge creation. For every month, except December, there were at least 500 messages inside communities, and 200 stories added to the environment. Because the Brazilian summer vacations are usually in the last two weeks of November and all weeks of December, it is expected a decrease on all collected metrics related to these months. But it is necessary to monitor these indicators for at least six more months in order to find a more accurate pattern.

7

ISBN: 978-972-8924-82-9 © 2009 IADIS

Looking to the number of added contacts, which is presented on the Figure 3, it was possible to verify that, in a span of six months, the number of new relationships varied from 385 in July 2008 to 65 new relationships in December 2008. Except for the last two months, there were more than two hundred new relationships created by the social network users. As it was verified, the numbers of new relationships for the last two months of the experiment were far lesser than the overall for the four months before. This behavior probably is also related to the Brazilian summer vacations, and also to some kind of stability on this number. This kind of stability is expected as the social network mature and the number of new members starts to decrease.

Figure 3. Number of new Added Contacts

Figure 4. Number of new read items

Observing the variation of the number of read items (Figure 4), it is possible to verify a really active community. The number of read topics and stories varied from 3000 to 4000 monthly, which is more than 15 times the number of created stories and topics during the same span of time. It was also noted a decreasing on the number of read items for the last two months of the year, as expected. It is important to notice that the number of downloaded objects do not suffered on the last two months. A more deep analysis would be necessary to understand why this happened. These data also suggests that the objects do not take part on the day-to-day activities as does the topics and stories. But even with a high reading number, the number of rated items – that is, the number of topics and stories evaluated by users with a rate from one to five stars (Figure 5) – was not so encouraging. It was noted that an average of only 10% of users who read an item usually rate it. After some interviewing, it was identified that only 20% of interviewed users did know about the existence of the rating functionality or how they could benefit from it, which indicates that there is a problem related to the functionality usability. This behavior of not rating items also can be verified on the number of items with a positive average rate (Figure 6). Only a few topics and stories created did receive rates greater than or equal to three stars. This number is usually around 10% of all items created monthly.

Figure 5. Number of qualifications

Figure 6. Number of positively rated items

But far more encouraging than the number of qualifications or the number of positively rated items is the number of classified items (Figure 7). The created stories have an average of approximately 2.1 tags for each story. This number is even higher when applied to objects, which presents an average of 2.9 tags for each object. Even topics have an average of 1.3 tags for each created topic. These numbers also indicates a certain level of awareness of the importance of knowledge classification, so a piece of knowledge could be more easily found by another user who would need this knowledge. Besides a really high number of knowledge classifications, users do not take full advantage of this classification. Even with almost 500 different tags, the number of accesses to a tag, listing items tagged with a keyword (Figure 8) is really low.

8


Figure 8. Number of accesses to a tag Figure 7. Number of classified items

When looking to the numbers of items classifieds and confronting them to the number of access to the classification lists (tags and tagcloud), is possible to verify that users do care about classification, but apparently do not know how to access this classified item through the classifications mechanisms, which indicates that there is a problem on the classifications mechanisms, more precisely related to the interface for accessing the classified content. Another interesting indicator is the number of unique visitors to a.m.i.g.o.s. Because users are not obligated to use it, the number of unique visitors is not near the number of employees. C.E.S.A.R has approximately 600 employees, but a little more than 50% of them are regular users of the WBSN, even that more than 90% of them has logged in at least once. These numbers suggests that it still is necessary to bring and keep more users inside the WBSN environment. When interviewed, some users affirmed that sometimes they have trouble accessing the a.m.i.g.o.s because this tool is not present yet on their day-to-day activities. That kind of behavior is really an eye-opener, suggesting that probable the most appealing characteristic of an environment like that is to be part of the members’ day-to-day activities inside their companies. After this second phase it was possible to identify that the WBSN approach to KM is still working well, but metrics indicates the need for enhancements in certain aspects of the WBSN, like the classification mechanism, the rating mechanism and new functionalities which would bring users’ day-to-day activities inside the environment.

5. CONCLUSION AND FUTURE WORKS A large number of companies are choosing a WBSN approach on their knowledge management strategies. This work presented the use of a WBSN called a.m.i.g.o.s as the main tool for communication and knowledge management in a Brazilian software development company: C.E.S.A.R. The main goal of this work is to present and evaluate an alternative to align individual interests in an organizational knowledge management initiative, turning it a natural process. The a.m.i.g.o.s WBSN has been used by C.E.S.A.R since October 2006 and the results indicated that this approach has been well-succeed to give cooperation, collaboration and knowledge sharing support in C.E.S.A.R projects (Costa et al., 2008). Using a WBSN the organization adds some human-oriented perspectives on a traditionally system-oriented approach, which makes the use of WBSN as a KM tool quite promising, decreasing the gap to a dynamic approach to KM. Also according to the data collected during the second phase of this case study, four main improvements shall be made to the system in order to better attend users’ needs. The first one is the development of a new feature related to day-to-day activities. Because C.E.S.A.R is a project-oriented company, one of the best ways to support these day-to-day activities is related to project support inside the WBSN. With that kind of support, every project would have a special environment which would support its activities, and where every project member should access in order to keep up to date with the project status, discussions, documents, stories, etc. The second improvement is related to better exploit the classification mechanisms provide by the folksonomy module. According to the collected metrics, users usually classify knowledge with tags, but they

9

ISBN: 978-972-8924-82-9 © 2009 IADIS

do not use them when looking for information. Based on this behavior it is clear that the folksonomy module shall be modified in order to facilitate the access to the classified content. A third improvement is related to document management capabilities. The knowledge creation metrics shows little improvement and little activity regarding the uploaded files and bookmarked sites, which together constitutes the user objects. The development of a version control module allowing the uploaded files to have its versions managed would probably lead to a larger knowledge base, bringing files that without this feature would be placed outside the WBSN. Some improvements to the objects user interface probably would also lead to increasing document accesses. The fourth major improvement shall be made regarding recommendations. The data presented in this work indicates a modest number of both, manual and automatic recommendations. However, with a larger knowledge base, the automatic recommendation probably would have more relevant data to work on. In general this work can be used as a reference to other initiatives focusing on the use of WBSN as the main tool in a KM approach. The metrics defined are real and have been monitored for six months, setting some parameters regarding the KM efficiency of a WBSN approach in an environment like that on C.E.S.A.R.

ACKNOWLEDGEMENT This work was supported by the National Institute of Science and Technology for Software Engineering (INES1), funded by CNPq and FACEPE, grants 573964/2008-4 and APQ-1037-1.03/08.

REFERENCES Ahmed, P., Kwang, L., Mohamed, Z., 1999. Measurement practice for knowledge management in Journal of Workplace Learning: Employee Counselling Today, Volume 11, Number 8. pp 304-311. Bose, R., 2004. Knowledge management metrics. Industrial Management & Data Systems, 104(6), 457-468. Choi, B., Lee, H., 2003. An empirical investigation of KM styles and their effect on corporate performance. Information & Management, 40(5), 403-417. Costa, R. A. et al., 2008. A.M.I.G.O.S: Using Social Networks to Manage Corporate Knowledge. In IADIS International Conference Web Based Communities. Amsterdam, Netherlands. Davenport, T., Prusak, L, 1998. Conhecimento Empresarial: Como as Organizações Gerenciam seu Capital Intelectual. Rio de Janeiro: Campus, 256 p. Domingos, P., Richardson, M., 2001. Mining the network value of customers in Proceedings of the seventh ACM SIGKDD international conference on Knowledge discovery and data mining, p. 57-66. Iacobucci, D., Wasserman, S., Faust, K., 1994. Social Network Analysis: methods and applications. Cambridge University Press. Liebowitz, J., Suen, C., 2000. “Developing knowledge management metrics for measuring intellectual capital” in Journal of Intellectual Capital, Volume 1. Number 1. pp 54-67. Nonaka, I., Takeuchi, H., 1995. The Knowledge Creating Company. Oxford University Press, New York. Staab, S. et al., 2005. Social Networks Applied. IEEE Intelligent Systems, 20(1), 80-93. Wenger, E., 2004. Knowledge management as a doughnut. Ivey Business Journal, 68(3). Available at: http://www.iveybusinessjournal.com/article.asp?intArticle_ID=465 [Accessed April 12, 2009].

1

http://www.ines.org.br

10


“BECAUSE I KNOW YOU”: ELEMENTS OF TRUST AMONG PSEUDONYMOUS BLOGGERS Vanessa P. Dennen Florida State University MC 4453, EPLS, Tallahassee, FL USA 32306

ABSTRACT In order to develop a sense of community among online discussants, such as bloggers, it is necessary to have an underlying trust. This study examines how trust is manifest within a largely pseudonymous blogging community. Findings show that risk, reciprocity, reputation, affiliation, prolonged engagement and norms all impact trust and thus community relationships. Bloggers readily distinguish between the reader audience they know and desire to interact with and a potentially unintended audience that makes them feel unsafe with online self-disclosure, and their sense of trust fluctuates as they become more experienced within the community and various blog-based events unfold. Over time, sufficient trust develops among participants to allow them to shed pseudonyms and share their legal identities in more private venues. KEYWORDS Blog, community of practice, trust, identity, virtual community

1. INTRODUCTION Trust is an important issue to be addressed with regards to the Internet since it is a medium known for allowing people to present themselves as something or someone they are not, whether for good or bad or even merely for entertainment. While Internet-based interactions can be quite positive and either support existing or help create new relationships, they also can challenge an individual’s notion of privacy. Many people who engage in online communities simultaneously enjoy acts of personal sharing and self-disclosure (both on their own behalf and that of other people) and are wary of unknown others who may be deceitful or somehow cause them harm. So what happens when a group of otherwise unfamiliar people meet each other online? Do they forge relationships and share information, or monitor each other suspiciously? The answer to this question is dependent on context. This paper explores this issue of trust and relationship building among a group of bloggers who did not know each other prior to meeting online. This topic is important as we enter an age where an increasing number of human relationships are being formed online. Online forums allow people to find connections with others from the comfort of their homes and it may be easier for them to find specific groups or types of people than it is in the physical world. To this point, Karlsson (2007), in a survey study of Chinese American diary weblogs sites, found that readers seek out sameness when choosing what blogs to read. Additionally, many people are seeking support in online forums specifically because they can do so in a manner that is removed from their everyday lives and because they can find likeminded others with whom to interact. For example, patients may look online for others who share their experiences and, when they find them, wish to interact in a way that maintains some distance and privacy from their legal identity (Josefsson, 2005; Orgad, 2005). In other instances, people are engaged in social networking, but may wish to keep their social lives separated from their professional lives and reputations. Regardless of the specific circumstances, this phenomenon of congregating and communicating online continues to grow. Understanding the parameters that govern trust as it relates to relationship development is a critical part of creating online communities that are safe places for people to

11

ISBN: 978-972-8924-82-9 © 2009 IADIS

interact regardless of topic and also will help identify reasons why some online communities thrive or fail in the realm of relationship development.

1.1 Trust and Online Interactions Trust is a topic that concerns a variety of online communities and their transactions. Perhaps most commonly discussed is the issue of trust in e-commerce commerce-related communities such as eBay, in which transaction partners must trust each other to either provide goods as described or pay for those goods in a timely and reliable manner (Boyd, 2002). Another trust concern relates to information and whether web sites are trustworthy in their content. This issue is particularly contentious with regards to sites like Wikipedia, which rely on a community of contributors to submit trustworthy additions to articles. The dynamic nature of Wikipedia’s content with updates from a potentially huge pool of authors with unknown credibility makes trustworthiness of its content, and by proxy its contributors, difficult to ascertain with any true accuracy from moment to moment (Dondio, Barrett, Weber, & Seigneur, 2006). Wikipedia relies on author reputation as well as the archived revision history measures that may indicate the trustworthiness of a particular article. Algorithms have been developed to generate numerical measures for the trustworthiness of text (Adler, et al., 2007). In other situations, system users provide ratings of a member’s contributions, such as the usefulness ratings for amazon.com book and product reviews. These indicators of peer support ideally can be used to gauge the trustworthiness of a given reviewer. Although it is possible to create computational models that measure trust in online settings based on factors such as reciprocity (Mui, Mohtashemi, & Halberstadt, 2002), the utility of these models remains in a narrow domain. They may be able to measure the accuracy of factual content on a site like Wikipedia or predict the likelihood that a transaction will be a positive one on a site like Amazon.com or eBay, but they cannot gauge the gut-level feelings of trust (or lack thereof) that develop among long-term participants in an online environment. Trust is necessary in an online forum in order to achieve a civil society and a true sense of community, and persistent identity on the Internet is one of the key tenets of such trust (Jordan, Hauser, & Foster, 2003). Indeed, it is through community that norms are established and through the violation of norms that we may note distrust (Joinson & Dietz-Uhler, 2002). Seeing the opposite of the phenomenon being studied can be useful for establishing the conditions of its presence, particularly when the phenomenon, like trust, tends to be unspoken in many interactions and simply results in smooth interpersonal relations. Although some people may use pseudonyms when interacting online, identity can be developed via consistent use of the same pseudonym along with at least a modicum of self-disclosure (Dennen, 2006). Readers note a consistent voice and details that fit together and from those can develop a sense of the person behind a pseudonym. Although creating new online identities is a simple endeavor, newcomers to an online forum must earn trust via a demonstrable persistent identity. With persistence comes a sense of accountability, too (Friedman & Resnick, 2001). People are concerned with the credibility of information sources on the Internet (Briggs, Burford, Angeli, & Lynch, 2002), and the sense of credibility extends down to individuals with whom they might interact. Much as Lincoln and Guba (1985) recommend prolonged engagement as a way of establishing research credibility, it is a way of establishing credibility among people who interact online.

1.2 Research Questions and Study Context This naturalistic study addresses the following research questions: What factors contribute to the development and sustenance of trust in a social blog-based community? What impact does this trust have on interpersonal relationships? And what happens in instances where trust is lacking? An ethnographic case study approach is used to examine a community of bloggers. This is a longitudinal study, as I first began interacting with this community in 2004 and began formally studying the community in 2005. The community being studied, Blogademe (see kfluff (2006) for the origins of this name), consists of bloggers who identify as academics in a range of positions (graduate student, postdoc, professor, administrator, etc.). They are an informal and self-selected community, consisting of individually authored blogs connected by weblinks and comments. They are bound by a common interest in discussing events in their professional and personal lives. The bloggers, with a few exceptions, write under pseudonyms and did

12


not have any connections to their fellow bloggers in the physical world prior to engaging with the community.

2. METHODOLOGY The data for this study were collected via a combination of interviews and observations. Interviews were conducted via telephone or Skype with forty bloggers over a period of one year and typically lasted between 30 and 90 minutes. A semi-structured protocol was used and the same topics were covered in all interviews. Bloggers were asked to describe their blogs, tell the story of their blogging experience to date, comment on their blog identity and practices, share how they perceive and feel about the blogging community, and comment on intersections between their blogging world and physical world. The length of each interview varied based on the blogger’s available time and verbosity. Observations were conducted over a two-year period during which I subscribed to RSS feeds for over 200 blogs. The exact number of blogs followed at any time varied, as some blogs were closed during that time period and others were started or discovered. Field notes were used to document events and incidents that crossed the community and appeared on multiple blogs as well as to note general trends and the overall feeling within the community at different points in time. Relevant posts and comments were archived for analysis. Data analysis involved coding the archived posts and comments for indicators of trust or distrust and the impact of this trust on the overall sense of community and then noting the emergent trends or patterns that transcended multiple blogs. The interview data served as a point of clarification, providing additional insight into the motivations behind particular interactions and practices noted on the blogs. Data were triangulated by type, source, and theory.

3. FINDINGS: ELEMENTS OF TRUST 3.1 Risk Trust exists when we knowingly take risks and think we will emerge unharmed. For many of these bloggers, there is a feeling of risk in the mere act of having a blog. Critics of blogs have suggested that they will not be hired or might lose their jobs because of the blog, as if the act of having that kind of online space means that someone also will use it for a tell-all exposé about their colleagues and institution or will neglect their scholarship in order to write on the blog instead (Tribble, 2005a, 2005b). There has been one well-known instance in which a faculty member blogging pseudonymously as Phantom Professor (phantomprof.blogspot.com) wrote about her life as a writing profession and included somewhat unflattering descriptions of students and their actions. When her real identity was discovered by the university there were ill feelings and media attention (Inside Higher Education, 2005). She was not invited back to teach at the university during subsequent terms. While Phantom Professor was not a part of the community her story is widely known and her tale serves as a cautionary one to those who choose to blog in this vein. Phantom Professor’s experience is an example of risk based on the actions of community outsiders. In other words, it represents what might happen if the wrong person reads the blog. Other potential consequences of blogging are being embarrassed or harming one’s relationships in the physical world via a post that has come across the web browser of the wrong person. For example, some bloggers have written about people in their lives and do not want those people to read what they have written. Alternately, they may simply not want certain people who know them in another context to be privy to what they write about on their blogs. The risk based on community insiders would be that a fellow blogger discovers and reveals a blogger’s legal identity or shares the blog with someone the blogger knows in their physical world. This risk seems to be of much lesser concern to the bloggers, who have an unspoken code of conduct that would prohibit such behavior. They trust that others who engage in similar risks would follow the golden rule of “do unto others as you would have other do unto you” and further that their fellow bloggers have some level of vested interest in keeping the community collegial.

13

ISBN: 978-972-8924-82-9 © 2009 IADIS

Ultimately, the risk that they take by blogging reflects Solove’s (2007) thesis that what one publishes on the Internet may take on a life of its own and lead to unintended negative consequences for oneself or someone else. Still, the bloggers willingly engage take these risks – admittedly, adjusting their words and level of disclosure to match their own comfort levels – which demonstrates both that the activity contributes value to their lives and that they have some fundamental level of trust in their audience. This finding, in turn, begs the question of who their audience actually is and who they perceive them to be.

Evidence of Risk Risk – real or perceived – is made evident on blogs via both proactive and reactive behaviors relating to the transparency of one’s identity. Proactively bloggers may refer to things that are unbloggable, cull certain sensitive posts from the archives, and edit existing posts. All of these actions are intended to reduce one’s vulnerability should the blog be found. Evidence of these actions are made visible within the text of posts with bloggers commenting reflexively on what they will not blog or why they have removed or edited something. For example, posts that have been edited may be replaced with the word “poof” or a similar pithy message and perhaps an indicator of why the post has been edited. Reactive behaviors occur once a blogger perceives or knows that their comfort zone has been breached and include acts of flight, such as shutting down a blog; damage control, such as removing posts, editing posts, or disabling archives; and contrition, such as posting a mea culpa post. After a reactive behavior, bloggers most often reassess their sense of risk within the community and may adjust their blogging habits accordingly. For example, some bloggers will stop posting about certain sensitive or vulnerable topics altogether while others may take a brief hiatus from blogging until they feel safe again.

Monitoring Risk These bloggers do not walk blindly into a potentially risky situation. Many use site statistics, such as sitemeter (www.sitemeter.com), collected behind the scenes to monitor who is visiting their blogs and what entries these visitors read. When bloggers notice that they are getting heavier than usual traffic due to a number of hits from a particular site, they may feel there is a risk that their blog is too visible to people outside the community. Further, when they notice hits from IP addresses at their own university or in their hometown or that someone at a particular IP address is systematically reviewing their archives they may perceive that they are at greater risk than normal. Additionally, many bloggers use commenting systems that log IP addresses and they like being able to identify the location of their commenters in certain instances, particularly when those commenters are otherwise unknown and make negative statements.

Reducing Risk The biggest risk to a pseudonymous blogger is identification by readers – whether long-term readers, casual passersby, or people from their physical worlds – and a resulting negative consequence. Anticipated negative consequences include embarrassment on the mild end and loss of job or relationships on the more serious end. Further, bloggers risk the loss of a safe place to write and interact with others, an effective silencing on certain topics depending on the audience involved (e.g. family, colleagues). There are a number of actions that bloggers can take to reduce their perceived risk of being outed. One is to prevent their blog from being indexed in search engines. This action means that in order to find the blog someone must have the URL or follow a link from another web site. Either way, it is likely that people who find the blog are already somewhat attached to the Blogademe network (e.g., a regular reader of other community blogs who follows a blogroll link) and that relationship adds somewhat to a sense of security. When staying out of search engines seems insufficient and having one’s thoughts out on the web in a publicly accessible area feels risky, bloggers may start to use password protection for some or all of their blog posts or they may begin a second blog which is private. In this instance, one needs to be established in order to have an existing readership to invite to the second blog. At this point, the issue of determining who is an acceptable or trusted reader is raised. This form of risk reduction comes with a definite trade-off because it closes off the possibility of being found by new bloggers and makes it more difficult for a blogger to be remembered within the community. Thus, a blogger must leave a bigger footprint via comments elsewhere in order to be engaged. Additionally, given a high number of bloggers who use RSS feeds, a technology that does not work for password protected blogs, password-protected blogs tend to further experience lower overall traffic.

14


3.2 Prolonged Engagement Prolonged engagement is one of the factors that contribute to a sense of trust within Blogademe for two reasons. First, there is an identity issue related to prolonged engagement with particular individuals. Given that the participants are interacting with pseudonyms and constructed identities, bloggers feel a greater sense of trust when interacting with people who have well-established blogs and pseudonyms than they do with those who are new to the community. In other words, a fellow blogger is generally considered more trustworthy than a first or second time commenter because she has an established commitment to the community and has demonstrated a consistent identity over time. This identity-related trust matters when a blogger is determining whose advice to follow – not that an established blogger would automatically be trusted – and with whom additional, off-blog self-disclosure might be desirable. For example, at times offers are made to read and provide feedback on a manuscript or to send a present or book. These interactions would involve sharing one’s legal identity and generally are not undertaken lightly unless a blogger feels she “knows” the person making the offer. Further, prolonged engagement helps a blogger know who is and is not aware of and following community norms, which are discussed below. Similarly, the bloggers tend to distinguish new or occasional readers from long-term ones and are more likely to trust the ones who have been present for a while. Although short of going to password protection they cannot control who does or does not read their blogs, they can use shortcuts in their posts that would speak directly to the long-term audience and leave the more casual readers confused, particularly when there are no archives available for finding a back story. Second, prolonged engagement within the community in general promotes a greater sense of what might transpire among participants. This situational trust, based on what bloggers have observed across multiple blogs and in many interactions, contributes to a sense of trust or distrust of the medium. For example, there are bloggers who have been found by students or colleagues and the community is aware of what has happened as a result. Based on these vicarious experiences, individuals may feel more or less safe and may decide to engage in more or less self-disclosure.

3.3 Reputation, Reciprocity & Affiliation Reputation, reciprocity, and affiliation are three separate constructs, but they are clearly interrelated in Blogademe and thus will be discussed together. There is a saying that one is judged by the company one keeps, and it holds true when applied to weblog reputation and affiliations. In other words, one’s affiliations contribute to their reputation – although it should be noted that there must be a sense of reciprocity in the affiliation. Bloggers indicate their affiliations by linking to, mentioning, and commenting on other blogs. In this context, it is not sufficient for an individual to create links to and make comments on blogs but rather the individual must, in turn, be recognized by those same blogs and bloggers. Additionally, it is necessary to differentiate between affiliation connections that simply demonstrate awareness, such as links to blogs that one reads with no expectations of reciprocity, and connections that are in the spirit of community. Within Blogademe, one way that newcomers can become accepted and gain trust of others quickly is through affiliation with a more established member of the community. When the old-timer adds a sidebar link or, making a stronger case for acceptance, mentions a newly started or discovered blog in a post, others are likely to follow the link. Similarly, if a person regularly appears as a commenter on an established blog and especially if they are acknowledged with replies they start to be considered part of the community. Reputation-based trust is, of course, more complex to establish than just through affiliation, but once a blogger is generally accepted as one who belongs, reputation can be developed. Key elements of a trustworthy reputation include a consistent voice over time (trust of the pseudonym/identity), and respectful interactions with others (trust that one cares about the well-being of the community), and reciprocity that goes beyond providing a link. Reciprocity is a critical component to the development of community within Blogademe. Without it, the blogs would exist in isolation, but with it they form a set of individually owned platforms that are shaped through their affiliation and interactions. Individuals give to the community with the tacit assumption that they also will be able to somehow take or benefit from it. Giving and taking occur in the form of advice or support and are found in both posts and comments. Bloggers may outright ask for advice or receive it unsolicited, and also tend to leave posts and comments for others that indicate they think of each other’s

15

ISBN: 978-972-8924-82-9 © 2009 IADIS

interest and needs. It is not unusual to see a post or comment that took considerable time to research or write that is explicitly for the benefit of a particular community member. Bloggers share their experiences and resources freely with each other. Essentially, the bloggers demonstrate trust via their reliance on the pay it forward principle, helping each other out with faith that someone else in the community will provide help in their hour of need. Trust and reciprocity also relate to off-blog identity and story sharing. When contacting each other via email, bloggers frequently reveal their legal names, knowing that additional identity elements can be found with a simple web search. If one blogger signs an email with a legal name, the other is likely to include her legal name in the reply. The same is done with other identity elements. This act is a demonstration of mutual trust and provides a sense of safety (i.e., if you “out” me, I can “out” you). However, as noted above, prolonged engagement is related to the willingness to engage in this reciprocity. A one-time commenter might send an email to a blogger and not get much self-disclosure in the reply. These elements all had a direct impact on the success of this research. During interviews, multiple bloggers shared that they had been contacted by other individuals to participate in research but did not do so unless they felt their identities would be protected. Either probed or unprobed, these individuals indicated that their willingness to participate in this study was, as one blogger stated, “because I know you.” Knowing the researcher, in this instance, meant that I was familiar through my interactions with them and others in the Blogademe blogs as well as through my own blog.

3.4 Norms Norms serve as a bed of common understanding in Blogademe. They help define who is in the community and who is not. These bloggers have set expectations of how others will react to their posts and comments. For example, if someone posts happy news then congratulations are anticipated and if someone posts sad or frustrating news sympathy is expected. Philosophical disagreements are handled respectfully, with the same tone as a calm face-to-face discussion and with a sense of agreeing to disagree when necessary. Insults and name calling, which occur commonly in some Internet forums, are not tolerated here, and personal diatribes are relegated to one’s own blog rather than the comments of someone else’s blog. These norms make Blogademe a safe place for bloggers to share some fairly intimate feelings and stories and get feedback and advice from others. Another norm related to trust in Blogademe focuses on the handling of personal information. When personal information is known, either because of off-blog contact, an identity slip on the behalf of the blogger (e.g., accidentally typing a real name or commenting from a legal identity account), or behind the scenes curiosity sleuthing, it is not repeated. In fact, many bloggers feel the need to actively warn each other about potential breaches of privacy. These warnings may be done privately, via email, or publicly in comments (e.g., “You might want to remove your colleague’s name from the second paragraph”) and posts (e.g., a post about how weather indicators in the sidebar provide publicly accessible code that gives away the blogger’s location). Pseudonym use is an interesting community norm that impacts trust. Although outside critiques have likened pseudonymity to anonymity and the latter typically raises suspicion, in Blogademe people develop identities around pseudonyms. Although some bloggers use pseudonyms because of the sense of identity security they afford, others have adopted them because everyone else uses them. Whereas in some online communities where pseudonyms are not a well-established norm people may be suspicious of those who use assumed names, in Blogademe that suspicion is reserved for people who post under the generic shared name “anonymous” and who do not engage in a level of self-disclosure that builds a sense of identity around a name.

4. DISCUSSION Returning to the initial questions asked in this study, which asked what factors contribute to trust in Blogademe, what impact the trust has on interpersonal relationships and what happens when trust is lacking, it becomes evident that trust plays a major role in the daily events and relationships of this blog-based community. The use of pseudonyms is a factor that potentially could contribute to a lack of trust since it is

16


natural to be suspicious of one who masks their identity, but in practice people using established pseudonyms and maintaining consistent behavior prove themselves trustworthy. Trust develops when the bloggers mutually engage in a set of activities, including risk-taking, reciprocity, and prolonged engagement. Further, bloggers rely on cues such as following norms, having an established reputation and affiliation with known others to determine how much particular individuals should be trusted. Interpersonal relationships among the bloggers would not develop without some underlying sense of trust. It is this trust that helps Blogademe be a community rather than a mere collection of topically related blogs. That the bloggers often engage in off-blog interactions and share their legal identities is a major indicator that that the relationships developed on these blogs go beyond co-existence and mimic those we find in physical communities. Levels of trust and self-disclosure vary among participants, with norms that help define interactions on a continuum from undersharing to oversharing. Undersharing tends to result in fewer readers, commenters, and relationships whereas oversharing may generate readers and commenters but jeopardize trust. The overall level of self-disclosure in Blogademe can be rather high and unconventional, often extending to personal details that one’s colleagues and friends in the physical world may not know. This selfdisclosure is possible because of the trust among bloggers to not cross worlds unless invited to do so and the nature of the relationships among the bloggers which is simultaneously surprisingly intimate for a public forum and yet cloaked in a sense of respect for privacy. Generally distrust in Blogademe is exhibited in pockets – perhaps by one blogger who is feeling unsafe based on site statistics, or a group of bloggers who do not trust a particular commenter or doubt the authenticity of a pseudonym. Although the community has experienced outings, these typically have occurred without malicious intent. On a few occasions outsiders have threatened to out a blogger, and in these circumstances the blogger has recognized that having a pseudonymous identity is not quite as precious and critical as initially thought. These occasions have led various bloggers to knowingly loosen their own precautions, changing their blogging practices as necessary avoid potential embarrassment, and soldier forward, sometimes more confident and secure than ever. Alternately, those who cannot find a comfortable spot again will leave the community because for them the risk is not worth the gain.

5. CONCLUSION These findings demonstrate that trust frequently develops via prolonged engagement in an online community, confirming findings within other pseudonymous (e.g., Dibbel, 1998) and non-pseudonymous (e.g., Rheingold, 2000) online communities. Although the blogs in this study represent one community with a particular area of common interest, the broad themes related to trust seem likely to hold constant across other, similarly structured communities regardless of focus. Risk, reciprocity, and prolonged engagement, which have proven to be key elements of trust in Blogademe, all have been attached to the research and theory literature on trust. The twist here is that this trust is being formed via a medium in which people cannot see each other and do not know each other’s legal identity. This study helps show how self-disclosure is critical to the development of online community and is calculated by participants to walk the fine line between acceptable and unnecessary risk. Further, it shows that trust is not a static construct within an online world. Instead, bloggers may redefine their comfort zones daily based on community events and perceptions of risk.

ACKNOWLEDGEMENT I would like to thank the many bloggers who so graciously allowed me to interview them and shared their community.

17

ISBN: 978-972-8924-82-9 © 2009 IADIS

REFERENCES Adler, B. T., Benterou, J., Chatterjee, K., Alfaro, L. d., Pye, I., & Raman., V. (2007). Assigning trust to wikipedia content. Santa Cruz: University of California. Boyd, J. (2002). In community we trust: Onine security communication at eBay. Journal of Computer Mediated Communication, 7(3). Briggs, P., Burford, B., Angeli, A. D., & Lynch, P. (2002). Trust in online advice. Soc. Sci. Comput. Rev., 20(3), 321332. Dennen, V. P. (2006). Blogademe: How a group of academics formed and normed an online community of practice. Paper presented at the Fourth International Conference on Multimedia and Information and Communication Technologies in Education, Seville, Spain. Dibbel, J. (1998). My tiny life: Crime and passion in a virtual world. New York: Henry Holt. Dondio, P., Barrett, S., Weber, S., & Seigneur, J. (2006). Extracting Trust from Domain Analysis: A Case Study on the Wikipedia Project Autonomic and Trusted Computing (pp. 362-373). Friedman, E., & Resnick, P. (2001). The Social Cost of Cheap Pseudonyms. Journal of Economics and Management Strategy, 10(2), 173-199. Inside Higher Education (2005, May 11). The Phantom Professor. Inside Higher Education Retrieved March 30, 2009, from http://www.insidehighered.com/news/2005/05/11/phantom Joinson, A. N., & Dietz-Uhler, B. (2002). Explanations for the Perpetration of and Reactions to Deception in a Virtual Community. Soc. Sci. Comput. Rev., 20, 275-289. Jordan, K., Hauser, J., & Foster, S. (2003). The Augmented Social Network: Building identity into the next-generation Internet. First Monday, 8(8). Josefsson, U. (2005). Coping with illness online: the case of patients’ online communities. The Information Society, 21, 141-153. Karlsson, L. (2007). Desperately Seeking Sameness. Feminist Media Studies, 7(2), 137 - 153. kfluff (2006). Tiny circle of friends Retrieved October 22, 2006, from http://kulturfluff.blogspot.com/2006/09/tiny-circleof-friends.html Lincoln, Y., & Guba, E. (1985). Naturalistic Inquiry. Thousand Oaks, CA: Sage. Mui, L., Mohtashemi, M., & Halberstadt, A. (2002). A Computational Model of Trust and Reputation. Paper presented at the 35th Hawaii International Conference on System Sciences. Orgad, S. (2005). Storytelling online : Talking breast cancer on the Internet. New York: Peter Lang. Rheingold, H. R. (2000). The virtual community: Homesteading on the electronic frontier. Cambridge: MIT Press. Solove, D. J. (2007). The future of reputation: Gossip, rumor, and privacy on the Internet. New Haven: Yale University Press. Tribble, I. (2005a). Bloggers need not apply. The Chronicle of Higher Education Retrieved March 16, 2006, from http://chronicle.com/weekly/v51/i44/44c00301.htm. Tribble, I. (2005b). They shoot messengers, don't they? . The Chronicle of Higher Education Retrieved March 16, 2006, from http://chronicle.com/weekly/v52/i02/02c00201.htm

18


FOSTERING STUDENTS’ LEARNING OF COURSE MANAGEMENT SYSTEMS BY USING BLENDED LEARNING Dr. Ahmed El-Gamal

Prof. Dr. Thomas Köhler

Menofia University, Egypt Educational Technology Department (DAAD post-doctoral scholarship holder at Technische Universität Dresden)

Technische Universität Dresden, Germany Fakultät Erziehungswissenschaften Professur für Bildungstechnologie (Education Sciences Department -Educational Technology Chair)

ABSTRACT Course management systems CMS, particularly Moodle, gained a growing interest in education at all levels because of the wide rang of resources and activities they can offer. Therefore, it is becoming essential for future educators to master their use and learn how to employ them in educational settings efficiently. This study aims to investigate the role of blended learning in fostering students’ learning of course management systems within the context of teacher education in Egypt. In comparison experiences from German universities are reflected. According to this lecture, workshop and webbased methods are employed to provide students with rich learning experiences that promote their learning and support their building of online learning communities. These methods are also used to accommodate students’ different learning styles and in turn enable their construction of knowledge and skills. This is a quasi-experimental study that evaluated students’ use of CMS and investigated their views of blended learning. The results reveal that blended learning enhanced the building of online learning communities; supported students’ learning of CMS; and promoted their interdependence, participation, interactions, and creativity. KEYWORDS Blended Learning – Course Management System – Moodle – Learning Styles

1. STUDY CONTEXT AND BACKGROUND The Ministry of Higher Education in Egypt started a national project in 2005 for the integration of information and communication technology in higher education institutions (ICTP). E-learning is considered to be the core of this project as it aims to enhance the quality of the student learning experience, promote greater and wider access to the university's courses, and hence to improve the effectiveness of teaching. In order to implement this project, an e-learning centre within each Egyptian university has been established in order to promote the use of e-learning through development of online courses and staff training. Since 2008 Egypt and Germany collaborate on researching eLearning technology under the frame of a joint programme of the science ministries. However, a number of challenges are facing the use of e-learning in Egyptian higher education system such as the need to change the dominant role of the teacher (mainly as lecturer) to become a guide and facilitator of students’ learning (Fetherston, 2001). There is also a need to change the role of the student from passive receiver of information to an active learner, in addition to the need to encourage students to collaboratively work in groups online. One of the barriers is also the lack of institutionbased online learning environments that both teachers and students can use. Subsequently, there has been a growing interest in Egypt to use open source software (such as Moodle) to support students’ online learning. In Germany the largest academic installation of an open source software based Virtual Learning Environment (VLE) is the OPAL system in Saxony with approximately 49.000 students, the additionally existing Moodle user group contains approximately 16.000 students (data from spring 2009; cp. Schwendel and Fischer, 2009). The theoretical background for this study is the social-constructivism approach to learning which is influenced by Vygotsky as this study gives particular attention to the social process of interaction and

19

ISBN: 978-972-8924-82-9 © 2009 IADIS

participation and the way in which individuals can construct shared knowledge (Salomon, 1997). Therefore, students’ online group work is emphasised and the ways that can support students to build online learning communities in order to enable their exchange of knowledge and ideas (Bridglall, 2001; Kahnwald and Köhler, 2007). Ryba et al. (2002) propose two basic principles that underpin learning communities whether online or face-to-face. The first principle is ‘commonality’ which involves a process of working together in common areas and interests and, in the process, forming a bond with one another and with the group as a whole. The second principle is ‘interdependence’ which implies depending on one another in a positive way for information, knowledge organisation, and shared problem solving. These principles draw attention to the importance of forming groups of students that share the same interests and are capable of self-direction of their learning. The educational context for this research is teacher education in Egypt, and in particular educational technology studies. This research had been implemented as part of a course entitled ‘distance learning technologies’. Within this course, the use of course management systems is introduced to the students in order to enable them to use such systems in distance education successfully. According to Moodle official site (http://moodle.org/), Moodle is a course management system (CMS), also known as a learning management system (LMS) or a virtual learning environment (VLE). Moodle as an open source course management system has been used in this study for a number of reasons. First of all, it is developed in the light of the social-constructivism theory and thus it can support the building of online learning communities. Secondly, it is an open source software, which is freely available and therefore, we would not face any budgetary constrains. Thirdly, it has many advantages such as its rapidly growing number of users, developers, and its online support community (Koohang and Harman, 2005). Accordingly, Moodle is also used to create powerful, flexible, and engaging online learning experiences (Rice 2006). Accordingly, Moodle has provided this study with a wide range of learning resources and activities. According to Rice (ibid), Moodle provides three different types of materials and activities, which are: 1. Static course material that a student reads, but does not interact with. These materials include a text page, a web page, a link to a file and a view into one of the course's directories. 2. Interactive course material that a student interacts with, by answering questions, entering text, or uploading files. These materials include assignment (uploading files to be reviewed by the teacher and/or students), choice (a single question), journal, quiz (an online test) and survey. 3. Activities where students interact with each other which are used to create social interactions. These activities include chat (live online chat between students), forum (by establishing online bulletin boards for each course), glossary (students and/or teachers can contribute terms to site-wide glossaries) and wiki (wikis can be inserted into courses, or a wiki can be the entire course). This study aims to use a blended learning approach to foster students’ learning of CMS and to adapt with the students’ different learning styles. Accordingly, a systematic approach has been used in order to effectively integrate various learning techniques, technologies, and delivery modalities to meet specific communication, knowledge sharing, and information needs and in order to overcome the limitations of faceto-face and online education (So and Brush, 2008; El-Deghaidy and Nouby, 2008; Owston et al., 2008). Blended learning brings traditional physical classes with elements of virtual education together (Akkoyunlu and Soylu, 2008). While, e-learning environments ensure the flexibility and efficacy which cannot be found in a classroom environment, face-to-face learning environment provides the social interaction which is required for learning. Blended learning has been also used in this study to improve teaching and learning outcomes; to increase flexibility and access to knowledge; and to support social interaction (Graham, 2004; Owsron et al., 2008). However, there are a number of issues which are relevant to designing blended learning systems (Graham, 2004). The issues of the role of live interaction; the role of learner choice and self-regulation; models for support and training; and cultural adaptation have been taken into account. Consequently, a systematic approach based on project management has been used in order to ensure the successful use of blended learning in this study (Bielawski and Metcalf 2003). Accordingly, two key aspects have been taken into consideration: 1. Selecting appropriate delivery methods for specific learning outcomes, and 2. Effectively combining diverse learning events and interventions into a holistic training and development program based on a blended learning model. For that purpose, we used a mix of inter-related methods which included lectures, workshops and webbased learning. These methods not only differ in the ways in which they are used, but they also differ in the

20


underpinning models that support them. For example, while lectures represent a teacher-centred model, workshops and web-based learning mainly represent teacher-plus-student and student-centred models. The use of such diverse models within the framework of blended learning aims to gradually shift the pedagogy from teacher-centred to become mainly student-centred (El-Gamal and Hudson 2001). Subsequently, adapting pedagogies to students’ learning styles can enable students to gain control of the learning situation.

2. RESEARCH METHODOLOGY The research methodology used in this research is a quasi-experimental study methodology. Such has been defined by Cohen et al. (2007) as a research methodology in which the researcher undertakes its study with groups that are intact, i.e. the groups have been constituted by means other than random selection. Among the different forms of quasi-experimental they mentioned, this study can be best described as the one-group posttest only design methodology. It includes one experimental group of students and, at the end of the course, a test on CMS in order to identify students’ progress. In addition, a questionnaire for using blended learning to teach the use of course management systems has been developed in order to identify students’ views about the use of blended learning. This questionnaire included 22 items that are based on Likert scale and ranged between 3 and 5 options for students to choose from. The questionnaire was divided into 4 main sections according to the different methods that were blended in this study. These sections include: lectures, workshops, web-based and blended learning. Both face validity and content validity of the questionnaire were considered by giving it to a trialling panel of educational experts. The questionnaire is reliable as the reliability coefficient (cronbach’s Alpha) is 0.681. A learning styles inventory had been also applied in order to identify the different learning styles of the students and whether the blended learning approach can accommodate these styles or not. For that purpose, the learning styles inventory which is based on auditory-visual-kinesthetic was used (Dunn et al., 1995). Furthermore, students’ activities on Moodle had been analysed in order to find out the level of their participation and engagement in the online course management system. This study included 200 students who are in their final year of educational technology studies at Menofia University in Egypt. Research tools (questionnaires and learning styles inventory) were distributed on all students. The total number of students who completed both the questionnaire and the learning styles inventory is 171, 58.5% are females and 41.5% males and their ages ranged between 20 to 24 years. The analysis of students’ learning styles shows a variety which included visual, auditory and kinaesthetic styles, but also mixed learning styles. While the majority of students (114 out of 171) prefer the visual learning style, a small percentage of them prefer mixed learning styles such as auditory-visual, auditorykinesthetic and visual-kinesthetic. The cross-tabulation analysis between students’ gender and their learning styles did not show any significant different between male and female students’ preferred learning styles.

3. DATA ANALYSIS AND RESULTS 3.1 Lecture Method Lectures have been used in order to orient the students with the subject of course management systems and their underpinning theoretical background and also to provide them with the basics for its use. Lectures also aimed to deliver essential knowledge to the students which is related to the use of course management systems. The analysis of students’ views revealed that the majority of students (91%) found that lectures did provide them with the essential knowledge and skills for using course management systems. In addition, most of the students (81%) felt that the orientation and the introduction to course management systems were sufficient. On the other hand, the majority of students (57%) did not feel that the problems they face during the course can be solved during lectures to some extent (see table 2). A large percentage of them (39%) were not sure that they can interact well with the teacher during lectures, while 30% of them did not think having had the opportunity to express their opinions during lectures because of a limited time.

21

ISBN: 978-972-8924-82-9 © 2009 IADIS

Table 1. Data analysis of students’ views of lectures.

Items

Values

There is an opportunity for me to interact with the teacher during lectures. I had the chance to ask questions and express my opinions during the lectures. I feel that the problems that I face during the course can be solved during lectures.

Yes

To some Extent

No

61% 64% 43%

39% 30% 50%

---6% 7%

These results indicate that the use of lectures alone can not fulfil students’ expectations, as they can be used for providing strong background about the subject matter only, rather than facilitating interaction and problem solving. Therefore, lecture as a teaching method, with its limitations, needs to be used alongside other methods that do better support social interactions and problem solving.

3.2 Hands-on Workshops Workshops inside the computer lab have been used in order to provide students with hands-on experience of using course management systems. This aims to enable the students to understand and manage the different tools that are available in a course management system and learn how to employ these tools in education. While lectures were open to all 200 students, workshops were organised in 8 smaller groups of students (each had an average of 25 students). Working in smaller groups during workshops supported more interaction among students and between students and the teacher. According to table 3, the largest proportion of students (89%) believed that working in small groups inside the computer lab helped them to develop their skills. This emphasises the role of face-to-face interaction when allowing students working together in small groups. In addition, 82% of the students confirmed having gained basic skills for using course management systems through the workshops. Table 2. Data analysis of students’ view of workshops.

Items

Values Strongly Strongly Agree Not Sure Disagree Agree Disagree

I feel confident about my ability to use e-learning management systems as a result of my study. Through workshops, I gained the basic skills for using e-learning management systems. I can now manage an electronic course as a result of my study. Working in small groups inside the computer lab helped me to develop my skills.

25%

52%

14%

9%

----

23%

59%

13%

5%

----

28%

50%

17%

5%

----

42%

47%

6%

5%

----

The results also revealed that the majority of students found that working in workshops developed their ability and their confidence in using course management systems. Moreover, workshops enabled them to manage online courses effectively (see table 3).

3.3 Web-based Learning The approach to use the web in teaching and learning has been adapted from Salmon's (2002) model. Accordingly, the online teaching and learning within a course passes different stages that are highlighted in her model, such as access and motivation, on-line socialisation, information giving and receiving, knowledge construction and development. In our study, the use of web-based learning seeks to provide the students with a type of learning that suits their individual learning needs and pace. As students can manage their own learning without feeling limited to any time and/or place this type of learning does also facilitate discussion and conversations between students through communication channels available - which in turn can support the building of online

22


communities. As at the beginning of the course, students formed 21 groups of shared interest, and created 21 online courses on Moodle and all courses were related to their field of study (educational technology). Within these courses, students acted as tutors and students at the same time because they created their own online courses to demonstrate their abilities in using course management systems. Figure (1) shows a screenshot of one of the online courses created by students in Moodle.

Figure 1. An example of one of the courses created by students on Moodle.

The analysis of results shows that 60% of the students faced difficulties in connecting to the Internet to some extent. However, the majority of the students (94%) appreciated the freedom to learn any time and any place. According to table 4, a large portion of students (71%) valued the freedom to present their ideas and suggestions to others through the Internet. Table 3. Data analysis of students’ views toward web-based learning.

Items The Internet and the course management system (Moodle) facilitate the discussions and conversations between me and my peers. I can freely present my ideas and suggestions to the others through the Internet. Learning through the web helped me to build channels of communication between me and my peers and between me and the teachers. I did not face any difficulties to connect to the Internet during the course.

Values Yes

To some Extent

No

71%

28%

1%

71%

26%

3%

71%

20%

9%

40%

43%

17%

These results also confirm the role of the course management system (here Moodle) in facilitating discussions and conversations among students. They also emphasise that students’ learning through the web helped them to build channels of communication with other students and also with the teachers.

3.4 Students’ Views of Blended Learning Finally students’ views on the use of a blended learning approach have been investigated. The results show that the use of the three teaching methods (lectures, workshops and web-based learning) is considered being necessary by the students. They found each of the three teaching methods complementing each other. For example, a large percentage of students (43%) found that the use of lectures and workshops compensated the difficulties they faced in connecting to the Internet. This confirms the fact that each of the three teaching methods has its own capacities and limits and therefore, they can be used together to provide an added value.

23

ISBN: 978-972-8924-82-9 © 2009 IADIS

According to figure (2), the majority of students (54%) found that the three methods (lectures, workshops and web-based learning) were utilised appropriately in the context of distance learning technologies. In addition, they felt that the three methods were integrated in a way that each one of them played an important role and can not be eliminated.

Figure 2. Students’ views of blending the three teaching methods.

The results also show that students’ preferences of a certain teaching method in relation to their learning styles varied to some extent. As shown in figure (3), the three teaching methods can be ranked according to students’ preferences as follows: 1st is workshop with a percentage of (47%), 2nd is web-based learning with a percentage of (33%) and 3rd is lecture method with a percentage of (20%). In addition, the analysis of data did not show any significant relationship between students’ learning styles and their preferred teaching methods, as the spearman correlation coefficient is 0.001. This indicates that the use of blended learning was successful in accommodating the different learning styles of students, which is in line with the results by Wageeh and Hitendra (1999) but also Liu and Ginther (1999). However, the results indicate that the majority of students prefer the workshop teaching method, which in our case included face to face interaction with the teacher, practical use of Moodle inside the computer lab, and interaction with other students - which is itself a kind of blended learning.

Figure 3. Students’ preferences of teaching methods.

The integration between the teaching methods is seen to be the key to a successful use of blended learning, as each method should not be used isolated. Rather the methods should be an integral part of the whole process. Such an integration needs to be completed in a systematic and creative manner in order to employ methods that compensate each other without replicating the processes (So and Brush, 2008; ElDeghaidy and Nouby, 2008; Owston et al., 2008).

3.5 Analysis of Students’ Progress Students had been tested at the end of the course in the different aspects and skills that are related to the use of course management systems. The analysis of the test results has shown a significant progress in students’ performance, as their average score is 25.42 out of the test total score 30. The results also show that the majority of students had achieved more than 75%, which is according to the grading system ‘very good’ (see

24


figure 4). In addition, the results did not show any relation between students’ learning style and their results in the CMS test.

Figure 4. Analysis of students’ results from the CMS test.

Students’ performance on the test indicates that the blended learning led to progress in students’ performance, as their average score was more than 75% which is ranked according to the Egyptian grading system as very good. In addition, the analysis of data did not show any significant relationship between students’ learning styles and their performance on the test, as the spearman correlation coefficient is 0.087. This result indicates that the blended learning approach accommodated successfully the different learning styles of the students. Students’ performance was not only measured by the CMS test, but also through evaluation of students’ online activities. The evaluation of online courses created by students demonstrated their competency, competitiveness, interdependence and creativity. The evaluation also revealed that students succeeded to form online communities, as they manage their own online courses successfully in groups. In addition, students achieved the course objectives, as they demonstrated their ability to: - create online courses on Moodle and customise its user interface following their own motivation; - manage course participants in terms of adding, deleting and updating users’ profiles; - design and upload course materials; - create and manage different types of course activities such as quizzes, assignments etc.; - moderate and actively participate in discussions at different levels (unit, course and site levels); and - exchange their ideas, opinions and resources. Students’ online activities on Moodle have been analysed in order to identify their level of engagement and contribution to the course management system. The analysis did not rely on counting students’ logs into Moodle, rather it focused on analysing students’ authentic activities such as adding, deleting, updating and viewing of the different resources, messages and activities on Moodle. Table (5) highlights the wide range of students’ activities, which is extracted from the Moodle’s reports of the online courses created by students. Table 4. Analysis of students’ activities on Moodle.

Add

View

Update

Delete

N of Groups

21

21

21

21

Range

894

3683

558

191

Minimum

138

536

18

14

Maximum

1032

4219

576

205

435.90

1.983

157.43

60.29

Mean

Although the majority of students’ activities were viewing (which is expected), other types of activities, such as adding, updating and deleting, took place and demonstrated high-level of participation and contribution. The difference between the minimum and the maximum number of activities draws attention to the differences between the groups in terms of their online contributions and their activities on Moodle.

25

ISBN: 978-972-8924-82-9 © 2009 IADIS

4. CONCLUSIONS It can be concluded that the use of a blended learning approach in this study is successful in fostering students’ learning of course management systems. This study also concludes that the use of a blended learning approach meets students’ needs and expectations. The systematic integration between the three teaching methods (lecture, workshops and web-based learning) is seen to be the key to the success of using blended learning. Firstly, lectures provided the students with the theoretical background for CMS. Secondly, workshops provided more motivating and engaging learning experience for students. Finally, Web-based learning played an essential role in facilitating students’ interaction, communication and exchange of information and resources. It also enabled the students to build online learning communities in which the students shared the same interests and goals. The variety of tools, resources and activities that are provided through Moodle provoked students’ competitiveness; motivated their participation and interdependence; and promoted their creativity. The findings in the context of this research led to a new conceptual approach on how to build student online learning communities. It is hoped that a systematic development will indeed lead to an increased independent student learning activity within self chosen social structures of a community of practice type. Although we can not claim to make generalisation from this case study, but it can influence future research in terms of designing effective blended learning options.

REFERENCES Akkoyunlu, B., and Soylu, M. Y. 2008. A Study of Student’s Perceptions in a Blended Learning Environment Based on Different Learning Styles. Educational Technology & Society, 11 (1), 183-193. Bielawski, L. and Metcalf, D. 2003. Blended eLearning Integrating Knowledge,, Performance, Support, and Online Learning. Amherst, HRD Press Inc. Bridglall, B. L. 2001. "Research and Practice on How People Learn." Pedagogical Inquiry and Praxis 1: 1-4. Cohen, L. et al. 2007. Research Methods in Education. London, Routledge. Dunn, R. et al. 1995. A meta-analytic validation of the Dunn and Dunn model of learning style preferences. The Journal of Educational Research 88. 353–362. EL-Deghaidy, H. and Nouby, A. 2008. Effectiveness of a blended e-learning cooperative approach in an Egyptian teacher education programme. Computers & Education. 51. 988–1006. El-Gamal, A. and Hudson, B. 2001. Developing a Constructivist Internet Curriculum for Teachers. ICCE Conference, Korea. Fetherston, T. 2001. Pedagogical Challenges for the World Wide Web, Educational Technology Review. 9 (1). Graham, C. R. 2004. Blended Learning Systems: Definition, Current Trends, and Future Directions. In Bonk, C. J. and Graham, C. R. (Eds.). Handbook of blended learning: Global Perspectives, local designs. San Francisco, CA: Pfeiffer Publishing. Kahnwald, N. and Köhler, T. 2007. Microlearning in Virtual Communities of Practice? An explorative analysis of changing information behaviour. Micromedia & eLearning 2.0: Getting the Big Picture. Proceedings of Microlearning 06; Innsbruck, University Press. Koohang, A. and Harman, K. 2005. "Open Source: A Metaphor for E-Learning." Informing Science Journal 8. Owston, R. et al. 2008. Blended teacher professional development: A synthesis of three program evaluations. Internet and Higher Education 11. 201–210. Rice, W. H. 2006. Moodle: E-Learning Course Development. Birmingham, Packt Publishing. Ryba, K. et al. 2002. Analysing the effectiveness of on-line learning communities. Higher Education Research and Development Society of Australasia's (HERDSA). Perth, Western Australia. Salmon, G. 2002. E-tivities: The Key to Active Online Learning. London, Kogan Page. Salomon, G. 1997. Novel Constructivist Learning Environments and Novel Technologies: Some Issues to be Concerned with, European Association for Research on Learning and Instruction (EARLI) Meeting, Athens, Greece. Schwendel, J. and Fischer, H. 2009. E-Learning an sächsischen Hochschulen – Strukturen, Projekte, Einsatzszenarien; Dresden, TUDpress. So, H.-J. and Brush, T.A. 2008 Student perceptions of collaborative learning, social presence and satisfaction in a blended learning environment: Relationships and critical factors. Computers & Education. 51. 318–336.

26


SOCIAL LEARNING FOR UNIVERSITY STAFF Francis Brouns, Adriana J. Berlanga, Sibren Fetter, Marlies E. Bitter-Rijpkema, Jan M. van Bruggen, Peter B. Sloep Open Universiteit Nederland, CELSTEC (Centre for Learning Sciences and Technologies) PO Box 2960, 6401 DL Heerlen, The Netherlands

ABSTRACT Social learning is the utilization of social networks and social technology for specific organizational learning outcomes. In the university’s organization contexts these outcomes are, for instance, academic staff professionalization. In this paper we present an initial exploration on how a social Learning Network to enhance academic staff competences can be designed. The starting point is to explore the use of social networks sites (SNS) in the academic staff and, considering the results, suggest what characteristics a social learning network should have to enhance competence development. A survey on the use and appreciation of SNS was conducted among our university staff. The responses showed that although staff is familiar with SNS, the actual use and participation is much lower. Staff tend to browse and visit special purpose SNS, but register to general and professional SNS. Reasons to join and participate are mostly social: finding and connecting to people. In a social Learning Network for competence development, a clear aim, next to indicators of quality of participants and learning material was considered important. Motivations to learn include usability, appealing interface, clear expectations and learning support services. KEYWORDS Learning Network, Social network sites, professional development, lifelong learning

1. INTRODUCTION The current knowledge society requires a shift in education, not only towards a more demand-driven approach but also to one that takes account of the growing opportunities new media and technology offer. This shift in educational paradigm requires continuous professional development of staff, including teachers, researchers, administration and management. Staff needs to be aware of new developments in their areas, such as new didactical approaches, innovative technologies, and develop new and current competences to be able to apply this in their working activities (be it research, education or administration) and foster their careers. Social learning for university staff is referred to as the utilization of social networks and social technology for specific organizational learning outcomes as, for instance, professionalization of university staff. For an example in an organizational context see http://www.masie.com/social1. Learning Networks can cater for this. Learning Networks (LN) are online social networks where the participants organize their own learning process in line with their needs for competence development. The LN consists of people and learning resources organized around a particular domain or subject. The LN members can participate in the role of learner, teacher, as well as content provider. The learning resources or events vary from courses, workshops, to learning objects suited to assist the learner in obtaining the required competence. There are some examples and initiatives of so-called learning networks (e.g. the Scottish Social Services Learning Network, http://www.learningnetworks.org.uk/). However, these mostly consist of topdown organized traditional education supported with some form of online education and communication facilities. Our notion (Sloep et al., 2007) of LN as online social network sites takes this to another level. The LN are bottom-up self-organized networks for competence development in a particular domain. The LN are particularly suitable for non-formal learning as seen in professional development or learning on the job. Participants are self-directed: determine what, when and how they want to study, geared towards their own personal needs. Thus participants have to take control over their learning process and activities, not only by actively looking for competence development plans, learning actions and people to support them, but also actually contributing and sharing those with others. We observe similar phenomena in the popular online

27

ISBN: 978-972-8924-82-9 © 2009 IADIS

communities and social network sites that thrive on social interaction: people use these sites to find resources (both content and people) for their particular need at that moment. Our main assumption is that LN should self-organize into communities and that social interaction between its participants is paramount to foster knowledge transfer (i.e. exchange of information and thus learning). We investigate how to design LN, including required functionality, in such manner that social interactions are promoted and stimulated to enhance knowledge transfer and efficient and effective competence development. In this paper we focus on affordances for competence development that would be required when designing LN for professional development of university staff, including research, teaching, management and support staff in a Technology Enhanced Learning Centre. These staff needs to keep up to date with innovations in educational technology as well as new developments in web 2.0 in order to design their research, implementation and development of new technologies as well as use that in teaching (teach as they preach). Our hypothesis is that LN can be designed like SNS or by using existing social network sites, because SNS can foster the community aspects and thereby participation, contribution and knowledge exchange. SNS at least can indicate how to promote and enhance social interaction, contribution and exchange of information. In this paper therefore we investigate whether university staff is familiar with social network sites, whether they use them and what for. Next we asked them what affordances a social network site would need to have when its purpose would be competence or professional development. We start by briefly introducing LN and related work. Next, we present the results of a survey on SNS and affordances for competence development among academic, teaching, management and support staff of a Technology Enhanced Learning Centre. Finally we conclude with future work.

2. LEARNING NETWORKS There are many popular online communities or social network sites where people share all kinds of information. In this paper, we adhere to boyd’s (boyd and Ellison, 2007) definition of a social network site (SNS) as a web-based service that allows people to (1) construct a profile within a system, (2) have a list of contacts, network or friends with whom they share a connection, and (3) view and navigate through their connections and those made by others within the system. These sites show a high activity, and thrive on the social interaction between its members. This large social space, emerging from social interactions between the participants is required for social network sites to become effective: exchanging information or transferring knowledge, be that communicating with other people or exchanging content. However, interaction and knowledge exchange between people does not start automatically. Several behavioural, psychological and sociological theories are mentioned in literature that can provide some insights in people’s motives to contribute (Erickson and Kellogg, 2000, Millen and Patterson, 2002, Thibau and Kelly, 1959, Trivers, 1971). Ling et al. (2005) found an increase in contribution when the benefit became clear to the members; while Cheng and Vassileva (2005) applied several theories to increase participation and contribution in their educational system. Learning largely consist of interaction between people and exchange of knowledge. Learning is largely a social process where people learn by and from others. Learning also relies on transfer of knowledge from one person to another or via resources. This requires that people can find relevant other persons as well as resources. For effective LN, it is thus very important that the social space arises to stimulate effective knowledge exchange and the learning process. Next to social space, the characteristics of the network and the participants are important (Kester et al., 2006, Sloep et al., 2007, Berlanga et al., 2008). Three conditions underlie this social space and thereby the enhanced knowledge sharing: boundary (a clear and defined goal and rules to abide by), heterogeneity (varied community composition) and accountability (identify people and their actions and likelihood to meet again) (Kester et al., 2006, Sloep et al., 2007). These conditions can be found in most successful and sustainable online communities. In addition, successful SNS offer the participants, next to a clear goal, purpose and rules, functionalities to enforce sustainability. We characterize these as: self-management: facilities that allow members to create and manage their own presence as well as their contributions within the community; self-organization: facilities to allow and enhance interaction and collaboration between members; self-categorization: facilities to assist members in classification and evaluation of contributions; and self-regulation: facilities for members to control the level of privacy of contributions (Berlanga et al., 2007).

28


Our research focuses on investigating the factors that make SNS so popular and effective in establishing sociability and knowledge exchange, to see whether we can apply these to LN for professionalization, or design LN on top of existing social network sites. In this paper we present the results of survey we conducted among 102 academic (teaching and research), management, administration and ICT development staff in an expertise centre in the fields of Learning Sciences and Technology Enhanced Learning. The centre aims to research, develop and provide sustainable and evidence-based solutions for the advancement of learning at work, at school, at home and on the move. This is accomplished by combining state-of-the-art research in the Learning Sciences with the innovative powers of new media, mobile devices and the Internet. Research is organized around three topics: efficient arrangements of education and training, efficient methods & technologies for professional development, and innovative power of new media, mobile devices and ICT for more intensive learning experiences. The centre is involved in several project developing technologies and implementing research outcome and offers a Master in Learning Sciences. Among the respondents, 56.4% was female, and 43.6% male. Average age was 42.3 years (SD=11.167, min=25, max=61), but with a mode at 27 and another one at 47. The majority of the respondents worked in two of the major research programmes of the department: 29% in the arrangements of education and training programme which is more aimed at instructional design research and 29% in the technologies for professional development programme (a more educational technology research and development oriented programme). Most were involved in research (56.3%) or teaching (20.8). However in the more traditional, instructional design research programme, 82% was devoted to research and 12% to teaching, while in the more technology development oriented research programme, respondents mentioned a variety of activities: research still being the main activity, followed by ICT development, teaching, administration, and management. Over half (51.3%) held an MSc or comparable degree and 38.5% a PhD, with backgrounds in psychology, education or educational technology.

2.1 Social Network Sites SNS allow people to look for and find other people (new and existing) and exchange information, based on their profile and common interest. Many of these SNS thrive due to social interactions: finding previous classmates, friends, relatives and meeting new people. They started off in the personal context, but professional SNS are rising. These features of SNS can provide social help support services where participants are turning towards others in search for an answer to their question. Other SNS are formed around a particular purpose: sharing resources of varying nature. These SNS can be the source of relevant learning materials, but more importantly indicate what features and design principles encourage participants to upload and share content. All SNS have in common that they evolve around the members’ profile (although it is possible to browse the sites without having to register). The profile composition among SNS is fairly consistent, with a generic section and a more elaborate section for interest, education and “information about me” (Berlanga et al., 2008, Ellison et al., 2007, Gallant et al., 2007). Interestingly, nearly all SNS include a section about education and work in the profiles, even for those SNS that are mainly oriented towards friends’ networks. All SNS encourage members to interact with others and add as many contacts as possible by providing support and tooling for this (Berlanga et al., 2008). Our survey contained four questions to investigate what information the staff considered to be important in a user profile, either to present themselves or to get acquainted with others, and vice versa. Respondents indicated occupation, interest and expertise to be most important, followed by real name and school/university. They did not make a distinction between information in their own profile and that of others. The SNS are still increasing in popularity. For example, the Nielsen survey in 2008 (http://www.nielsenonline.com), indicate that MySpace is still the leading SNS, followed by Facebook, but that Twitter, Ning and LinkedIn are among the fastest growing SNS (based on US data). According to a press release of comScore (http://www.comscore.com/press/release.asp?press=2742) Facebook is the fastest growing SNS in Germany. Hyves is the most popular SNS among youngsters in the Netherlands. The respondents in our survey were certainly familiar with a lot of SNS, as indicated by the number of SNS they visited, but the actual use and participation as can be derived from the number of SNS they registered to, was much less. LinkedIn turned out to be the most visited and used SNS among our staff. The forty respondents indicated 311 times that they visited one or more of a total of 60 different SNS; with a range of two to 24. LinkedIn,

29

ISBN: 978-972-8924-82-9 © 2009 IADIS

YouTube, Plaxo, Flickr, and Hyves were among the most visited SNS. The list of 60 SNS seems impressive, but only 23 were mentioned more than twice. Most of the mentioned sites can be classified as special interest sites: sites intended for sharing all kinds of content; closely followed by general SNS (sites with as main aim connecting with friends). As the majority (27) of the respondents indicated research to be their main activity and only 10 teaching, it is not surprising that the research staff mentioned more (228) visited SNS, against 91 for the teaching staff. Both teaching and research staff however visited on average more of the general purpose SNS (Table 1). The majority of the respondents worked in either a more traditional educational research field or a more technology development research area. Staff in the first form of research visited less SNS (74), but more of the general (32%) SNS, while staff in the technology oriented research visited 111 SNS and more of the special interest SNS (37%) (Table 2). In the more traditional educational research group, 82% of respondents indicated research as their main activity and the remainder (12%) teaching. In the technology development oriented group, staff activities were 53% research, 13% teaching, 20% ICT development, 6% management, 6% administration. This could explain their more varied interest in types of SNS. Table 1. Average number of visited and registered sites for teaching and research staff, according to type of SNS

general SNS professional SNS special interest SNS total SNS cases

Teaching 4.3 (1.9) 1.7 (0.7) 3.8 (1.2) 91 35

Visited Research 3.2 ( 1.6) 1.8 (0.8) 2.9 ( 1.6) 228 107

Teaching 1.4 (0.5) 1.3 (0.5) 1.5 (0.7) 18 13

Registered Research 1.4 (0.7) 1.6 (0.5) 1.0 (0.0) 32 22

Table 2. Average number of visited and registered SNS by staff in a more traditional education research group or a more technology development oriented research group, according to type of SNS Visited Traditional research Mean (SD) Count general SNS professional SNS special interest SNS total SNS cases

2.6 (0.8) 24 1.5 (0.7) 12 1.9 (1.2) 21 74 45

Technology development Mean (SD) Count 3.7 (1.9) 33 1.8 (0.8) 18 3.7 (1.6) 41 111 44

Registered Traditional research Mean (SD) Count 1.2 (0.5) 5 1.3 (0.6) 4 1.0 (-) 1 10 8

Technology development Mean (SD) Count 2.5 (0.7) 5 1.3 (0.5) 5 2.0 (1.4) 4 16 9

However, the actual registration to a SNS was lower. We asked the respondents to indicate their first, second and third most frequently used SNS. Of the 40 respondents, only 29 (72%) had ever registered to a SNS, but 5 (12%) did not return to the SNS after registration. So, 28% never registered to a SNS at all. Although the respondents indicated that they visited general and special interest types of SNS, the SNS they registered to were the professional (career development) types, such as LinkedIn. LinkedIn was mentioned 11 times, Hyves 5, Facebook 4, and Plaxo 4 times. LinkedIn (7) and Plaxo (3) were used in work context, Hyves (5) and Facebook (3) in personal context. The staff in the more traditional research group seemed to prefer the professional and general SNS; while the staff of the technology oriented research group registered to general and special interest SNS (Table 2). The same SNS, Plaxo, MySpace, Facebook and LinkedIn, were among the sites people registered to, but stopped using. Reasons to stop using a SNS were: not meeting expectations, too time-consuming, low usability, no added value, better alternatives available and concerns around privacy. Similar factors were mentioned as drawbacks of SNS. The respondents to the survey mentioned social aspects (invited by friends or others, keeping in touch, meeting new people) as main reason to join, as well as the main reason to participate and the most appreciated characteristics. Respondents mentioned that they both acted as ‘listener’ and created or shared content. The respondents ranked the self-management functionalities add contacts, browse other people’s contacts and create a personal profile most often in first and second position. This corresponds to the findings by Ellison (Ellison et al., 2007) that Facebook promoted social capital, in particular bridging and maintained social capital. Bridging social capital enforces weak ties and allows users to build loose connections between

30


people. Maintained social capital allows users to maintain contact after disconnecting from a previous network (Ellison et al., 2007). Social interaction in networks or communities does not arise spontaneously. Online social question and answer sites rely on active participation of their visitors. This can be promoted by creating benefit for the participants, like the point system in Yahoo that favours those who contribute with high quality answers (Shah et al., 2008).

2.2 Affordances in Learning Networks In the current knowledge society and with the onset of technology and web 2.0, continuous competence development or professional development is required. Online communities like mailing lists and discussion forums, have been used for quite some time now both by professionals and individuals looking for an answer to a problem they face. Question-answer sites like Yahoo Answers, but also all kinds of support sites, like for example the communities around (open source) software applications, are popular and thriving. This, even informal and non-formal, process of obtaining answers to a question or problem constitutes a form of competence development. People tend to rely more and more of these forms of support instead of traditional curriculum based courses, because it is more flexible and geared towards their personal needs and requirements at that particular moment. It offers quicker options to arrive at a solution (King Research, 2007). LN should accommodate this by providing users with support for finding the right persons and resources for their needs and incentives or affordances to stimulate interaction and exchange. Our survey addressed this by asking several questions about what functionalities or affordances a SNS should contain to assist somebody in improving a competence in order to do their job better or obtain a better job position. In three open-ended questions we enquired what functionalities and features a competence development SNS should contain so that they would join, what the most important characteristic would be to assist in developing competences, what aspects or features would encourage learning.

2.2.1 Entice to Join Regarding the reasons to join a SNS intended for professionalization, respondents indicate the following features and functionalities they would like to have in order to sign up: • Social aspects: type of people present, experts, asynchronous and synchronous communication. • Expertise of participants: experts who can teach me something, people I know, people I trust, information about expertise and interests. • Quality of the learning material, like “learning resources and activities”, “worked out examples by experts”, “interesting course material”. • Learning support: assist in determining the goal, recommendations on people and resources, access to people and experts for support, communication facilities. Also, some people commented about a form of social support that involved other people in providing answers and advice; in few cases respondents indicated that the person should be an expert. • The boundary or purpose of the SNS: “a description of the competences”, “a list of goals, experiences, problems solved, … I could pick from”, clear goal, structure. • Openness of the system: ability to bring in own content, usability, employer benefits, but also being able to see in which other SNS a member participates, not using a dedicated SNS, but building on top of existing ones and including some fun.

2.2.2 Develop Competences Respondents indicate that learning material and content is the most important feature they consider for competence development. Particularly, staff ranked as most important characteristic: competence descriptions, sufficient learning material, identify learning resources and teachers, being able to organize and categorize content-related aspects, finding and contacting people and experts, and peers with a mutual goal sharing, and find out what are peers doing. As second most important characteristic, staff ranked mainly: • social aspects: expertise, how do I compare, list of people who reached competence, nice peers; • learning aspects: engaging self-tests, how do I compare, learning opportunities, simulations, list of people who reached competence. In addition, respondents indicated anonymous access, being able to browse, clear aim, reliability, transparency, mash-up with existing SNS, easy access: no costs, no (additional) registration, to be important

31

ISBN: 978-972-8924-82-9 © 2009 IADIS

2.2.3 Motivate to Learn Aspects or features that motivate to learn were found to be related to the boundary (clear expectations, functionalities, goals), and usability aspects (ease of use, reliability, user-friendly, appealing interface), more than to particular aspects of learning or competence development (trace learning paths, progress indicator, test skill development, the idea that there is something to learn). The sustainability functionalities included content as well as social aspects. For categorization and organization the social aspects (people with similar interests or activities, stimulating experts and peers, easily getting connected with peers, being able to distinguish between experts and newbie) were more important, for management the content (interesting and useful information, access to new information). Effectiveness indicators scored high (12) as well as usability and a dynamic and interactive environment (active participation, new information, stimulating experts and peers, interesting discussions, easily getting connected). But also personal benefit was mentioned (equal benefit, contribution, increased reputation. The respondents remained vague on requirements towards learning materials (interesting courses, learning opportunities), but were a little more specific on learning support (progress indicator, test skill development, trace learning paths, helpful solutions to problems).

2.2.4 Learning Support The respondents considered learning process related forms of support, like answers to content related topics, feedback, collaboration and advice on how to proceed the most important (Figure 1). Other forms mentioned include advice on materials to use, best path to follow; self-assessment and teaching. I don't know Other 5% 3%

Feedback 22%

Collaboration 21%

Other 14% Don't know 4%

Assessment 5%

Advice 19%

Question answering 25%

Choose from a list 7% Service suggests 4%

Combination 71%

Figure 1. Type of support considered most important (left) and preferred form of social support (right) (n=28)

Respondents clearly prefer a mix of methods of choosing potential people from a list and a service that suggests relevant people to be able to find suitable people to ask a question (Figure 1). Other forms of support mentioned, were: ask questions anonymous to a forum like feature; Google search for the site; have an expert answering my questions; only contact people that I know. The main reasons to help others were either for the good of the community (18) (altruistic, collegiality, community building, knowing the person) or because of personal benefit (17) (challenge, clear goal, benefit for self, get help in return, reputation, improving own skills), although quite a few (10) would help only when it did not involve any costs for themselves (available time, existing expertise). When asked to rank content and functionality a SNS can offer, self-study material was ranked highest, followed following courses and being able to contact experts (Table 3).

32


Table 3. Ranking of content and functionality Following courses/workshops/trainings/seminars Being able to contact other participants Being able to contact experts Being able to find participants that can help you Get a recommendation from the system who can help you Get a recommendation from other participants who can help you Relevant resources for self-study

Rank 1 6

Rank 2 7

Rank 3 2

Rank 4 4

Rank 5 -

Rank 6 2

Rank 7 4

total 25

3

3

2

6

3

4

4

25

6 2

6 2

8 5

1 4

4 9

2 -

2

27 24

-

3

3

4

3

5

7

25

-

2

4

2

2

9

5

24

11

5

3

4

2

1

1

27

28

28

27

25

23

23

23

177

3. CONCLUSION While the use of SNS for learners in formal and non formal contexts is a hot topic of research, the use of such technologies for professionalization of university staff is addressed only cursory. Although staff was familiar with the phenomena of SNS and visited quite a few SNS, actual use by registering and participating was much less. Most visited sites were the special interest sites, but respondents registered more to general and professional SNS. These general and professional SNS seem to be used for the social aspects mainly. Even when this is done in a professional context, it seems respondents mainly used the sites to find people (existing or new), used the sites for job advertisements or career change. The special interest sites like YouTube offer attractive and leisure resources that potentially could be used towards competence development. It is not clear whether respondents actually used the SNS for this, as the social aspects seem to outweigh the content related aspect. Even if the respondents did register to these sites, at the moment of the survey they did not consider these to be their most frequently used site. As the use of SNS is still growing, it is likely that also among the respondents the use and uptake of SNS has changed. Reasons to join, participate and main characteristics seemed to evolve around the personal aspects, indicating the importance of the accountability, self-management and self-organization aspects of the SNS. Respondents want to be able to identify and find people, based on a description or profile of the person, together with a qualification of that person, by expertise and interests. Similar findings are found in literature were the lack of face-to-face contact is often mentioned as the main obstacle in online learning (Stodel et al., 2006). The replies to the questions on affordances for a SNS directed towards competence development suggest that such a SNS could work. The SNS should provide the affordances that allow the participants to find the people they need, and the content they want. Not only being able to find these resources seems important, but also being able to find the correct or most suitable resources, as indicated by the need for categorization. A fair amount of the remarks dealt with the quality indicators, both of the participants and the content. The motivation to learn was influenced by the boundary and usability as well as incentives like effectiveness and personal benefit. Personal benefit can lie in a credit system as used in online social question and answer sites (Shah et al., 2008) or in reputation and effectiveness (Vassileva, 2008). The findings provide sufficient grounds to assume that a LN can be based on the principles found in SNS. A LN should accommodate the need for social interaction, e.g. by providing social support services to locate and find suitable people to address all kinds of questions to. The required profile information can be taken from existing SNS. Similarly, the LN should provide services to locate actual content, learning resources and learning activities which might be taken from existing SNS, although it is likely that this will be specifically addressed by the LN. Moreover, the LN should provide services to classify and organize people and resources, aggregated from several source SNS. A similar survey we conducted among psychology and computer science students indicated as well the usefulness of such a SNS in an academic setting. Currently we are preparing a prototype to investigate how a social learning network can be build using existing social

33

ISBN: 978-972-8924-82-9 © 2009 IADIS

networks and web 2.0 technologies (e.g., OpenSocial, mashups, etc.). The central idea is to take advantage of existing social technologies and benefit from them to enhance professionalization of academic staff in nonformal contexts.

ACKNOWLEDGEMENT The present work was carried out as part of the TENCompetence project, which is (partly) funded by the European Commission (IST-2004-02787) (http://www.tencompetence.org).

REFERENCES Berlanga, A., Sloep, P. B., Brouns, F., Rosmalen, P. V., Bitter-Rijpkema, M. E. & Koper, R. (2007) Functionality for learning networks: lessons learned from social web applications. ePortfolio 2007. Maastricht, the Netherlands. Berlanga, A. J., Bitter-Rijpkema, M., Brouns, F. & Sloep, P. B. (2008) On the importance of personal profiles to enhance social interaction in Learning Networks. IN Kommers, P. & Isaías, P. (Eds.) IADIS International Conference Web Based Communities 2008. Amsterdam, Netherlands, IADIS Press. boyd, D. M. & Ellison, N. B., 13(1), . http://jcmc.indiana.edu/vol13/issue1/boyd.ellison.html (2007) Social network sites: Definition, history, and scholarship. Journal of Computer-Mediated Communication, 13. Cheng, R. & Vassileva, J. (Eds.) (2005) User motivation and persuasion strategy for peer-to-peer communities. Ellison, N. B., Steinfield, C. & Lampe, C. (2007) The Benefits of Facebook "Friends:" Social Capital and College Students' Use of Online Social Network Sites. Journal of Computer-Mediated Communication, 12, 1143-1168. Erickson, T. & Kellogg, W. A. (2000) Social Translucence: An Approach to Designing Systems that Support Social Processes. ACM Transactions on Computer-Human Interaction (TOCHI). Special issue on human-computer interaction in the new millennium. Part 1, 7, 59-83. Gallant, L. M., Boone, G. M. & Heap, A. (2007) Five heuristics for designing and evaluating Web-based communities. First Monday, 12. Kester, L., Sloep, P., Brouns, F., Van Rosmalen, P., De Vries, F. & De Croock, M. (2006) Enhancing Social Interaction and Spreading Tutor Responsibilities in Bottom-Up Organized Learning Networks. IN Kommers, P., Isaías, P. & Goikoetxea, A. (Eds.) IADIS International Conference Web Based Communities 2006. San Sebastian, Spain, IADIS. Ling, K., Beenen, G., Ludford, P., Wang, X., Chang, K., Li, X., Cosley, D., Frankowski, D., Terveen, L., Rashid, A. M., Resnick, P. & Kraut, R. (2005) Using social psychology to motivate contributions to online communities. Journal of Computer-Mediated Communication. Millen, D. R. & Patterson, J. F. (2002) Stimulating social engagement in a community network. CSCW '02. New Orleans, Louisiana, USA, ACM Press, New York, NY. Research, K. (2007) The Value of Online Communities: A Survey of Technology Professionals. KACE. Shah, C., Oh, J. S. & Oh, S. (2008) Exploring characteristics and effects of user participation in online social Q&A sites. Sloep, P. B., Kester, L., Brouns, F., Van Rosmalen, P., De Vries, F., De Croock, M. & Koper, R. (2007) Ad Hoc Transient Communities to Enhance Social Interaction and Spread Tutor Responsibilities. IN USKOV, V. (Ed.) Sixth International Conference on Web-based Education WBE 2007. Chamonix, France, Acta Press. Stodel, E., J., Thompson, T. L. & Macdonald, C., J. (2006) Learners' Perspectives on what is Missing from Online Learning: Interpretations through the Community of Inquiry Framework. Thibau, J. W. & Kelly, H. H. (1959) The Social Psychology of Groups, New York, John Wiley and Sons. Trivers, R. L. (1971) The evolution of reciprocal altruism. Quarterly Review of Biology, 46, 33-57. Vassileva, J. (2008) Toward Social Learning Environments. IEEE Transactions on Learning Technologies, 1, 199-214.

34


PULLING DOWN THE WALLED GARDEN: TOWARDS A PARADIGM FOR DECENTRALIZED SOCIAL NETWORK MANAGEMENT Stefano Bortoli, Themis Palpanas, Paolo Bouquet Department of Information Engineering and Computer Science at University of Trento Via Sommarive 14, 38100 Povo (Italy)

ABSTRACT Social networking sites have gained much popularity in the recent years, because of the opportunities they give people to connect to each other in an easy and timely manner, and to exchange and share various kinds of information. However, these sites are based on a centralized paradigm, which limits the mobility of their users, and ultimately, their chances for establishing new relationships and benefiting from diverse networking services. In this paper, we argue for a decentralized paradigm for social networking, in which users retain control of their profiles, and social networking sites focus on the delivery of innovative and competitive services. Our position is that only in this environment will both the social networking sites and their users be able to develop to their full potential. KEYWORDS Social network portability, unique identification, web application

1. INTRODUCTION A social network is a structure composed by individuals or organizations, that are connected by one or more specific types of interdependency, such as values, ideas, friendship, etc.. In the past years, we have witnessed a fast growth of web based social network services aiming at building online communities of people willing to share their interests or activities [Freeman 2006]. During the early phase of social network service development, the focus was on connecting people tied by specific kind of relations, for example, ex-classmates. Recently, some social network services left this bind to specific relation between social-being offering a more general purpose social networking service. Furthermore, this kind of services allows people to build communities or groups of interest about specific topics, potentially satisfying the social networking needs of every potential user. This approach led to an impressive growth of few social network services involving tenth of millions of users often living in localized geographical area. Among others, the most successful web based social network services are Facebook and Myspace. This fast growth in the number of social network services users created a brand new business market related to commercial space and market analysis. This is one of the reasons that led social network providers to the definition policies aiming at reducing the “mobility” of users through the several social network service applications. In this context, we now want to move toward a model where individual users and their relations become the focus of social networking application. In this new model, users have more control over the data/information they own, such as the data that refer and are relevant to them, including the information relative to their social relations. Furthermore, the user should be empowered to perform some basic, yet fundamental, action on these data. An example is the capability of moving social networks information (i.e., the connection with other social network users) across different social network applications, keeping control over the relationship virtually stated. A first step towards this decentralized social network paradigm is to provide users with tools enabling them to define and edit their social network in a standard format and in a neutral environment. Another

35

ISBN: 978-972-8924-82-9 © 2009 IADIS

essential step to enable social network portability is the usage of an uniform identification system in the definition of social networks, boosting automatic integration when the user wants to access a social network application. In order to achieve these goals we have implemented a tool enabling users to manage personal social network profiles defined using the well known Friend Of A Friend ontology as standard vocabulary. Any social network profile presents only the FOAF description of a person and her social network (i.e. the list of known persons). Thereby, in order to assure an automatic social network integration with other social network profiles forming an extended social network graph we need to find a way to identify uniformly all the people across different profiles. To accomplish this objective we enabled our tool to interact with an Entity Name System (ENS) providing shared globally unique identifiers. This way, any user can define and manage her social network in a social network web application independent environment, using a standard vocabulary, and with the possibility of a smooth integration with other social networks based on person globally unique identification. Such social network profiles can in principle be used to access any social network application, making possible a concrete social network portability. The remaining of this paper is organized as follows. We first provide an analysis of other approaches aiming at social network portability (Section 2), we describe in details the foundations of the proposed paradigm (Sections 3 and 4), and we briefly present a neutral web based tool supporting users in editing social networks and integrate globally unique identifiers (section 4.1). In Section 5, we present the conclusions.

2. RELATED WORK Social network portability is an open discussion ongoing into the WWW community. In this section we present an overview of the known approaches aiming at making social networks distributed and portable. Tim O’Reilly introduces the concept of “social network operating system” in [O'Reilly 2007]. In this article the author foresees a system enabling the portability of social networks and personal information from social network application to others. O’Reilly points the uselessness of repeating the personal and social network description process any time a user access a new social network application. He aims at the definition of a set of user-centric application enabling the description of personal social network, focusing then on definition of tools enabling the automatic integration of the described information with other sources. This article presents mostly a vision and does not present any concrete solution. In the context of the Microformat initiative, a social network portability based on microformat is under definition (http://gmpg.org/xfn/). The proposed social network portability system, named Xhtml Friends Network (XFN), relies on hCard + xfn microformats for people and relations description. The social network description passes through HTTP references pointing at web pages (e.g. blog, personal pages etc.) of friends. Such web pages should then in turn embed xhtml hcard presenting the person description and further reference to friends. It is possible to notice how this approach replicates the idea of FOAF using an XHTML syntax rather than RDF. DataPortability (http://www.dataportability.org/) is a collaborative effort promoted by a group of volunteers and WWW marketers to enhance the capability of controlling, sharing, and moving data across independent systems. DataPortability endorses the idea that users should be able to move, share, and control for example their identity and all the relevant personal data as photos, videos, conversations etc. The project aims at defining best practices for integrating existing open standards and protocols. The main goal is to enable end-to-end data portability between online tools and services. This frictionless movement through the network of services should foster stronger relationships between people and services providers and help in building a healthy networked ecosystem. Open Data Definition (ODD - http://www.opendd.net/) initiative offers a framework supporting data portability across different social network applications. Basically, Open Data Definition provides a neutral format to represent people, sites, 'objects' and the relations among them. This format should then be used to implement export/import tool enabling the migration of social networks across the social network web application supporting it. The solution proposed seems valid from the technical point of view, involves high level concept of entities and relations between them. The main issue of this approach is that does not rely on

36


Semantic Web standard as RDF (Resource Description Framework) and deals only marginally with the entity identification problem. OpenSocial (http://www.opensocial.org/) is an initiative aiming at providing a set of common application programming interfaces (APIs) for web-based social network applications. These APIs are developed by Google along with MySpace and a number of other social networks providers. OpenSocial is commonly described as a more open cross-platform alternative to the Facebook platform. The main aim of OpenSocial is to create an ecosystem of social network application sharing common software APIs and templates. FOAF (an acronym of Friend of a Friend – http://foaf-project.org) is a machine-readable ontology describing persons, their activities and their relations to other people and objects. The FOAF project, which defines and extends the vocabulary of a FOAF profile, can be considered the first open standard for Social Semantic Web application, in that it combines RDF technology with ’Social Web’ concerns. A decentralization of social network system based on FOAF is proposed in [Yeung 2008 et al.]. In this paper, the authors sketch a system where users manage all social network information on a trusted server, relying on some access control system to enable social network application to access and use the “decentralized information” for social network activity. The central point of the paper is to give to the user the power to manage it's data and provide a central access point for social networking. The weak points of this approach are two. First the trusted server information repository must be able to guarantee access and manage all information needed to realize the social network services. This, in our opinion limits the evolution of social network services as, in order to access new services in one application, the user should set his social network information in a way that the social network application can use it to realize the offered service. Thus, any time a new service is offered, the central social network information repository must be updated in order to enable the access to such a new service. We believe that users should own the social network information, but that social network service provider should keep, respecting privacy issues, all the information necessary to provide specific services. Social network providers should compete on offered services in order to guarantee technological evolution. Second the social network integration relies on a weak identification system as URL pointing at FOAF profiles and inverse functional properties are volatile and do not guarantee complete social network integration [Bortoli et al. 2007]. The architecture of a platform for social network interoperability is described in [Mostarda et al. 2008]. In this paper the authors present the description of a software architecture enabling interoperability between different social network web applications. The work relies on two main bricks: 1) the user has an “identity page” presenting all the social network information according to the favorite standard, the identity page URL works as identifier for the user; 2) OpenId as single sign system to access all the social network applications. The platform would then provide connectors and converters enabling all the social network application to access the information in the user identity page according to the required format. In other words, the authors propose a sort of centralized “social network information” repository providing proper API supporting social network application with secure access to the user data. This project presents solution to many practical problems, but let unresolved the problem of identification of person within the user social network. The above work does not present any specific solution to the automatic social network integration problem once the user information is imported into the social network application.

3. PRELIMINARIES We now give a brief overview of the ENS (a more detailed presentation can be found elsewhere [Bouquet et al. 2008]), which we will use as the basis for our discussion. Note however, that our discussion is relevant to any system for entity identification management. The overall goal of the ENS is to handle the process of assigning and managing unique identifiers for entities in the WWW. These identifiers are global, with the purpose of consistently identifying a specific entity across system boundaries, regardless of the place in which references to this entity may appear. The ENS has a repository for storing entity identifiers (note that this repository will be distributed and replicated) long with some small amount of descriptive information for each entity. The purpose of storing this information is to use it for discriminating among entities, not exhaustively describing them. Entities are described by a number of attribute-value pairs, where the attribute names and the potential values are userdefined (arbitrary) strings. Clients interact with the system through the Access services layer. Clients can be

37

ISBN: 978-972-8924-82-9 © 2009 IADIS

both human users and applications, and may inquire about the identifier of an entity by providing a set of attributes that describes this entity. If the entity exists in the repository, the system returns its identifier. Clients may also modify the state of the repository, either by inserting a new entity in the system, in which case the ENS returns the newly assigned identifier, or by changing some of the attributes of n existing entity. As shown in Figure 1, the end result is that all instances of the same entity (i.e., mentioned n different systems, ontologies, web pages, etc.) are assigned the same identifier. Therefore, joining these documents and merging their information becomes a much more simple and effective process than before.

Figure 1. Schematic of the ENS and its interactions.

4. TOWARDS A DECENTRALIZED PARADIGM In this section we propose a new paradigm for social networks and their portability. The first point to analyze is with regards to the social network ownership. We believe that a social network (i.e., the set of connection between the user and the tied people) only belongs to the person describing it. When a user defines a new relationship with an unknown person in a social network application environment, the user should be the only owner of this information. Stating the contrary is comparable to constraining the realization of social activity only in the “place” where the two persons got in contact for the first time. For this reason we think it’s essential to define an open standard allowing the description of social network of a person enabling its complete portability. The social network information should not be constrained to the application in which the social relations between social-being is expressed. Users should be able to describe in a machineprocessable way their own social network only once, and in a single document under their complete control. The intuition behind our approach can be explained through a simple analogy: Imagine you go to a pub, and meet new friends. Even though the reason for meeting in that specific pub can be to drink a particular brand of beer, you and your friends are still the “owners” of your relationship, and such a relationship should be independent of the fact that you met in that pub. For example, one day you may decide to watch a football match with the same friends. Then you and your friends can simply meet at another pub that screens this particular football match. Hence, the two pubs stay the owners of the services they offer (serving good beer, or having a large screen for football matches), and they compete on the basis of the services they offer. Nevertheless, pub customers are not locked-in, which means that they can decide to use other services (or even the same services, but of better quality) in another pub. Social network information should be decentralized with respect to the social network application, and managed by the user as a private piece of knowledge. Any user should be allowed to manage/edit his/her social profile(s) out of any social network application. The user “loyalty” to a social network application should not due to the concerns about losing contact with tied people, but because the social network offers appealing services where to manage social relations. A view over a possible schema representing this new paradigm is depicted in Figure 2.

38


Figure 2. An overview over a possible structure of the decentralized social network paradigm

In this sense, what FOAF project proposes seems to be a suitable solution for the definition of decentralized and portable social networks. In fact, FOAF project is essentially about describing social network in a machine-processable way. Users are asked to define social network personal profile and make it publicly available as any other Web document. Unfortunately, this solution considered in isolation suffers from the following drawbacks. Lack of access control. Openly sharing social networks data raises privacy concerns related to information access control. Indeed, user might enjoy on one hand the advantages of social network portability, but on the other hand, they are not able to control which application is going to access their information. In our opinion this fact limits the adoption among average social network users, stressing again the private nature of social network information. Weak identification system. The use of FOAF ontology (as any other formally defined ontology) to define social network profile guarantees a uniform interpretation of the attribute describing the person and their connection. Unfortunately, FOAF suffers of some identification issue. In fact, the identification of persons in FOAF relies on the fact that different people would use the same values of specific kind of properties (i.e. inverse functional properties). It is quite natural that the values of such properties might not be shared uniformly among all the persons who want to refer to the same friend in their FOAF social network profiles. This decreases the concrete capability of automatically connect and integrate social networks [Bortoli et al. 2007]. Lack of Expressive Social Network Ontology. The FOAF ontology presents only the “knows” relation to tie two persons in a social network. This relation has the good property of being general enough to satisfy any kind of social network relation. Unfortunately, the capability of specifying the nature of the relation connecting two persons is quite important. It is actually the base of the definition of social network itself. Any subject is indeed part of different, probably disjoint, social networks defined according to the founding relations connecting him with other subjects. The fact that FOAF presents only one kind of relation prevents users to adopt it to specify different social network according to the context involving them. For example, its quite common to discern and keep separated work social network from friends social networks. Lack of Neutral User Friendly Tools. Along FOAF project lifetime many projects have been developed, but almost none of them was oriented to support end users in editing and keeping updated their FOAF profiles. In this way, also the people that created a FOAF profile had to manually edit the RDF code describing its social network, discouraging users in keeping up-to-date their profiles or even worst creating invalid RDF documents. We are convinced that the use of shared globally unique identifiers would solve the problem of an automatic social network integration. It is clear that a system like this must rely on a solid third-party identification system. Indeed, different users should be enabled to identify equally the same person across different private social network pieces of information. OKKAM ENS is the more general freely available naming service. Thereby, our solution for social network portability is the integration into FOAF personal profiles of globally unique identifiers for person provided through ENS interface. The owner of the FOAF social network profile should then be enabled to keep access control over the described information. In order

39

ISBN: 978-972-8924-82-9 © 2009 IADIS

to support users in managing FOAF profiles we defined a web based Rich Internet Application named foafO-matic. Such application is briefly described in section 4.1. It is important to notice how our approach is different from other identification approaches as the Single Sign On. In fact, our approach allows generically the global identification of persons in a complete application independent fashion. In principle, any application is enabled in recognizing whether two persons have been virtually identified in the same way. This is not possible using the Single Sign-On approach, since it only allows a specific user to use their identifiers for accessing some social network service application. Joining a social network application should consist of a simple operation: submit the social network profile to the social network application and accept its policy in terms of privacy management (e.g. information disclosure) and other related issues. The social network application should then recognize automatically all the people contained in the submitted social network profile that are already using the application and build the relative connections. The submission could be realized by uploading the social network profile or simply by providing a link to the document. The second option has the advantage that the user could keep up-to-date its social profile only in one location and any modification would affect any social network application. As previously mentioned, an undesirable side effect of this approach is that everyone would be enabled to access the openly the exposed social network information. A way to avoid this problem could be the creation of a sort of repository guaranteeing access control to the social network information. The social network information owner would then enable the social network application in accessing the information in a transaction on base of some safe protocol following the model of online payments. We believe this approach is good for users, because it allows them to move around web based social networking application more freely/easily, and consequently gives them more power and control over their data. Furthermore, the social network users are enabled in defining and, by using applications as foaf-Omatic, managing in a decentralized way their social network in a unique document they can use as a sort of “personal card” when accessing social network services. In principle, with this approach a user is also capable in defining different social network documents that can be used different kind of services, enabling mobility across all web based social network application offering a similar kind of services. For example, a user could describe only once the ex-classmates social network and use it to access the social network applications providing this kind of services. When social network applications enable the complete portability of social network s, several advantages come for social networking service providers. Indeed, if a user is free to move social network information across different platform, the “extended” social network brings in more users by minimizing the new user entry cost. Indeed, the user accessing the new social network application might want to invite in a batch all the friends that are using it, speeding up the process. Furthermore, even big social network service vendors can benefit of this approach. In fact, by allowing users to establish relationships across different social network sites, a wider variety of users with different interests/goals can be involved. This decentralized paradigm has advantages for both the social networking sites and their users (similar observations have also appeared in other studies [Ramakrishnan 2007]): 1. Data about users and their connections are under the control of the users, which means that they can in principle be uploaded into any social networking platform that is compatible with the format in which these data are structured. In this way, users can benefit from the different services offered by a variety of social networking platforms. 2. Social networking applications can focus on the task of providing value added services to users, and try to attract new users from a pool that is now much larger. By allowing users to establish relationships across different social network sites that cater to social networks with different interests and goals (e.g. artistic,professional, etc.), all social networking platforms can benefit from the increased mobility of users.

4.1 Foaf-O-Matic Foaf-O-matic is a Rich Internet Application based on J2EE AJAX framework enabling users to manage theirs social network profiles (see http://www.foaf-o-matic.org). A user is enabled to load and edit the personal social network profile, integrating globally unique identifiers provided through ENS.

40


As it is possible to see in Figure 3, the application allows to describe the profile primary person in the left part (marked with “Me”), and describe a list of friend in the right one (marked with “my friends”). The point 1 highlighted in Figure 3 presents the globally unique identifier for the person retrieved through ENS. Each friend, as the primary person, can be described with an arbitrary number of FOAF attributes (point 3 in Figure 3), and can integrate an ENS URI. The fact that a friend is globally unique identified is marked with a special logo in the friend description as highlighted in point 2 in Figure 3.

Figure 3. View of foaf-O-matic interface. The menu placed in the header of the application offers access to several functionalities: load FOAF profile from local machine and remote location; save FOAF profile on local machine and remote location via SFTP; visualize the generated RDF FOAF code; email FOAF profile; load friend and main person v-card; load friend's FOAF profile with facilities for importing friends of a friend description; SHA1 email address encryption; find friend’s globally unique identifier by searching in ENS; create globally unique identifier in the ENS; add and remove an arbitrary number of friends; edit friend description. The application is under constant evolution and we are looking forward to provide more services supporting users in managing their social network in a decentralized and “neutral” environment. Foaf-Omatic alone does not provide a complete solution to realize the decentralized social network paradigm described in section 4, but it's a first step towards the definition of an automatic decentralized social network integration system.

5. CONCLUSIONS In this paper, we presented an overview of the currently adopted approaches for enabling social networking, and argued in favor of web based communities that are decentralized and independent from social network web applications. We pointed out the limitations of the existing systems, and introduced a novel paradigm for the management of social networks in a decentralized and neutral environment. The work presented in this paper is the first step towards decentralized social networks, since it offers a viable solution to a central problem: globally unique identification. Foaf-O-matic is the first tool enabling the management of personal social network profiles outside any social network web application, using a machine readable standard vocabulary, and giving the possibility of uniformly identifying individuals across different

41

ISBN: 978-972-8924-82-9 © 2009 IADIS

social networks, through the interaction with ENS. We propose foaf-O-matic as the solution for two of problems affecting decentralized social networking based of FOAF, that are the lack of neutral user friendly tools, and the weakness of the currently adopted identification system. In order to complete the realization of the proposed paradigm, further steps are necessary, such as the definition of shared social network ontology. An extension of FOAF would be a promising direction to follow in order to accomplish this goal. This ontology should present attributes' metadata enabling the definition of access control policies limiting information disclosure. In this way, the user could set access control rules and safely submit her social network profile to the web application she likes more.

ACKNOWLEDGEMENT This work is partially supported by the FP7 EU Large-scale Integrating Project OKKAM – Enabling a Web of Entities (contract no. ICT-215032). For more details, visit http://www.okkam.org.

REFERENCES Linton C. Freeman, 2006, The Development of Social Network Analysis: A study in Sociology of Science. Empirical Press, Vancouver USA . Tim O'Reilly, 2007,The Social Network Operating System. Online article at http://radar.oreilly.com/archives/2007/10/social_network_operating_system.html Ching-man Au Yeung et all, 2008, Decentralization: The Future of Online Social Networking. W3C Workshop on the Future of Social Networking. 15-16 January 2009, Barcelona Bortoli, S. et all, 2007, Foaf-O-matic - Solving the Identity Problem in the FOAF Network. Proceeding of SWAP 2007 Fourth Italian Semantic Web Workshop. 18-20 December 2007 Bari Italy Paolo Bouquet and Heiko Stoermer, 2008, OKKAM: Enabling an Entity Name System for the Semantic Web. Proceedings of the I'ESA2008 Workshop on Semantic Interoperability. 01-05 June 2008. Tenerife, Spain. Ramakrishnan Raghu and Tomkins Andrew, 2007, Toward a PeopleWeb. IEEE Computer Society Press.Vol 40, No. 8, pp 63--72 Mostarda et all, 2008, Towards an OpenID-based solution to the Social Network Interoperability problem. W3C Workshop on the Future of Social Networking. 15-16 January 2009, Barcelona

42


SHARING EXPERIENCES USING WEB 2.0 AND KNOWLEDGE PORTALS FOR ADVANCING WEB COMMUNITIES Nik Bessis, Mitul Shukla and Zhihua Lai Department of Computing and Information Systems, University of Bedfordshire, Luton, UK

ABSTRACT The paper discusses the development of a web based knowledge sharing portal using intra and inter tagging functionality. This functionality enables multiple users to tag multiple documents in multiple ways at a macro and micro level. Users can upload documents and tag the whole document as well as parts there-in as they deem appropriate, other users can then tag the whole document or specific parts there-in as they deem appropriate. The flexibility offered by the technology allows for specific instances as well as specific types of information to be found across a given repository of knowledge. The technology also allows users to navigate across multiple, possibly disjointed, documents that are connected only in so far as they have been tagged using the same word or phrase by the user community. KEYWORDS Distributed Community of Practice, Folksonomy, Collaborative Tagging.

1. INTRODUCTION The paper discusses experiences learnt with the development of various web based knowledge sharing portals. These were the main deliverables from the Arts Council funded Ambition project and the JISC funded Users and Innovation programme in which authors have been involved with. These portals used an inter and intra tagging system developed at the University of Bedfordshire during 2007-2008 [1]. A synopsis of the user communities will be given followed by an overview of the methodological approach taken to develop the technology. Requirements capture and user needs identification will also be described. This will be followed by descriptions of service functionality and implementation, essentially how the technology works and the way that user communities could use it. Finally there will be an overview of the testing and evaluation procedures that were followed and a discussion of the results derived there from.

2. CASE SCENARIO Two communities were the recipients of the technology that was developed. The first were arts organisations taking part in the Ambition project. An aim of the Ambition project was to foster best practice in the use of ICT by arts and cultural organisations. A group of 15 organisations based in the North West and in the East of England participated. The second community comprised of participants of the JISC funded Users and Innovation (U&I) programme. The community was made up of 22 projects based within various higher educational institutions throughout England. The aim of this project was to investigate and implement the use of innovative practices and technologies within higher education establishments. Although the domains that these communities exist within do differ there were certain elements that both had in common. For example both communities were comprised of distinct independent groups of practitioners, both communities by the nature of the projects they were involved in were going through a learning experience and both communities were made up of groups that through their involvement and practice had produced experiential knowledge of their processes and outputs. Due to the nature of their dispersed characteristics both communities recognised the need for a web based knowledge sharing tool for

43

ISBN: 978-972-8924-82-9 © 2009 IADIS

much the same reasons as Hildrech, P. et al. observed “… improve organisational performance by maintaining implicit knowledge, helping the spread of new ideas and solutions, acting as a focus for innovation and driving organisational strategy.” [4]. With this in mind, the contributions of the paper are to: i) present our methodological approach; ii) discuss our service development; iii) including web portal implementation; and iv) offer relevant evaluation results.

2.1 Our Methodological Approach Initial investigation was with the arts/cultural organisations community as chronologically this group had a requirement first. The U&I community decided upon their need for a knowledge sharing portal later on. A User Needs and Innovation Development Framework (UNIDF) approach was employed. The approach synthesized Participatory Design and Rapid Application Development activities into various development stages. The approach taken appreciates that the user community as a Community of Practice (CoP) required the following levels of support: • The user community as a CoP made aware of the existence of the UNIDF approach and its process and benefits for them. • The user community as a CoP involved with using the UNIDF approach in practice. The approach used for assisting and enabling users to participate, network and develop cross-sector coherent implementations. • This level assumed a level of expertise amongst users so that they can use the UNIDF approach for criticism and for sustaining the user community by sharing experiences and innovations with a wider CoP. This also served as the vehicle to sustain user community and gather evidence for critical analysis and evaluation about our methodological approach. Figure 1 details the iterative process of the activities. Achiever's Iterative Process

Data From:

To produce:

Ambition:

Information

refine:

User Requirements:

Information

Reflection:

Information

to produce & refine:

to Identify:

specification

good practice in sector gaps public value

Taxonomy

filtered by content analysis

knowledge portal

content for case studies social networking tools

medium format for representing them

training material

requirements for knowledge base

produce: Aggregated case study (Stories)

about Arts Organizations

via Methods:

Knowledge Portal (outcome)

Portal Your Tube Blogs/Wiki Interviews Focus groups

Represented in various medium formats

Desk research

content viewpoints assessed

Workshops

Outcome key: 1st, 2nd & 3rd cycle

Logistics Director 2nd & 3rd cycle 3rd cycle

Manager

Arts Organizations

User types

consisting of users who want to view, adapt and comment on others' stories and ultimately tell their own stories! (I am the one, etc…)

Figure 1. The Iterative Process of our User Needs and Innovation Development Framework Approach (UINDF)

44


3. SERVICES DEVELOPMENT A Rapid Application Development approach was taken to the development of the knowledge sharing portal. Essentially the approach was an iterative process of building and testing with lessons learned or observations made from the testing phase being implemented in the next building phase. The research team conducted initial system testing in-house. The system was then previewed by members of the user community with further changes and enhancements being made to the system before it was trailed/critiqued by the wider community for further development.

3.1 Service Modelling Focus groups were employed involving arts organisation stakeholders. The aim of these focus groups was to create a baseline understanding of how the participants perceived themselves and their peers and their organisations in terms of experiential and learning change as well as the influence and use of ICT. Various elicitation techniques were employed in a manner inviting participants to consider experiences in the past, present and future. The elicitation was done with regard to the work experiences of participants. Participants were also provided with the opportunity to discuss these matters in an informal peer group setting. During the focus groups, participants were exposed to a case study describing the use of ICT by an arts organisation. Information was collected regarding its perceived usefulness in terms of what aspects of the case study participants found of interest to themselves as well as what they felt may have been lacking. This was useful in feed back to NMP, a consultancy whose task was to produce an initial batch of six case studies, which were to form the basis for us to observe the manner in which it would be beneficial for arts organisations and trigger them to prepare and publish their own stories and case studies using next generation technologies. A Grounded Theory [2] approach was adopted to content analysis of the output from participants. An important aspect of the results collated from the focus group content analysis was that participants wanted online case studies to contain navigation to other relevant case studies. Further, participants expressed views on specific aspects of the organisation, the nature of their job, what they thought and felt, and the work environment (social and physical). To achieve this content analysis of nine textual case studies was undertaken; three of these were developed by independent sources with the remainder being provided by NMP. The overarching aim was to identify key categories and produce a taxonomy of types of information that would inform the design of the knowledge sharing portal. The taxonomy served two related purposes; it allowed for the tracking and analysis of how users tag and search information within case studies, and; it helped to provide guidelines for users on how to tag and search information. An inductive grounded theory approach [2] was taken for this analysis. Each case study was analysed in terms of structure and content (types of statements) being made as well as supporting categorisations. Content analysis was also used with the data collected in the focus groups described earlier. The participants perceptions of organisational change and learning were identified, as well as, their experiences of using ICT. Seven organisations as a sample, from the arts/cultural community, were invited to participate in blogging about their experiences in the project to date. Participants were encouraged through emails to write about specific themes. They were also given the option to write about any other relevant topic, related to, or stemming from participation in the project. The blogs were set up to be confidential, only the participants and the project team had access to the content contained in them. It was hopped that through this means participants would express their views about what was happening in their organisations, reflectively. It was also an opportunity to capture and analyse the users reactions to the prototype of the knowledge sharing portal when it became available to them. These blogs could also help the organisations create their own case studies which they could upload to the knowledge sharing portal at a later stage.

3.2 Service Functionality The knowledge sharing portal was developed with a view to sit within a Joomla content management system as this adequately and securely dealt with issues such as user registration and logins as well as offering other tools for collaboration and community building such as fora and messaging. The knowledge sharing portal enables the annotation from differing perspectives to be applied to material on the same subject matter.

45

ISBN: 978-972-8924-82-9 © 2009 IADIS

Essentially the knowledge sharing portal allows multiple users to inter/intra tag instances of a given subject matter, see Figure 2.

Joomla Portal

Upload file

search for Tag1

Tag1 Tag2 Tag1 Tag4 Tag1 Tag1

Tag2 Tag3

Figure 2. Illustrative overview of the knowledge sharing portal.

Folksonomic tagging as a concept and a technology is fairly well used and understood (see http://del.icio.us) [3][5]. However the knowledge sharing portal differs from these systems by enabling multiple users to tag specific documents as well as specific components there-in. The functionality of the knowledge sharing portal thereby allows the searching of specific components within a given document and also across other documents that contain that specific tag. The use of user-generated tagging of documents allows for the most specific and relevant content to be identified and consumed by practitioners as opposed to software (search engines) or documents on a macro level (del.icio.us). At this stage any digital file type recognised by the PHP core of Joomla can be used on the knowledge sharing portal. All of these digital file types can be uploaded to the knowledge sharing portal and viewed via a browser in the normal way. The uploaded files can be categorised and sub-categorised. The knowledge sharing portal uses two levels of classification, categories and sub-categories. Categories work at a macro level and are used to cluster similar material. Category headings used for example were: Understanding the User, Technical Development or Audience Development. Sub-categories however work at the micro level within each of the aforementioned categories and can be helpful in disambiguating potentially large clusters of uploaded material in a given category. Sub-categories titles were: General Methods and Approaches, Specific Tools/Techniques and Outcomes/Benefits/Results. Categories and sub-categories were pre-defined and not user changeable. However, regardless of which category/sub-category a file was uploaded under, the user was able to specify the title and a brief description of the file during the process of uploading that file. The title and description were both searchable via the search field and were both examples of inter-tagging, that is, tags that describe the whole document. The knowledge sharing portal also uses intra-tagging, also known as micro-tagging. To date this facility is only available for text documents with the .txt extension and for image files such as jpeg, gif and png formats. Intra-tagging is essentially the user being able to select an area of interest within a document and labelling it from their own (subjective) perspective. This means that multiple users could tag the same elements within a given document using a variety of tag labels and thus making the content discoverable from multiple perspectives.

46


4. SERVICE IMPLEMENTATION Two versions of the knowledge sharing portal have been developed, one for each user community, Figure 3 illustrates the interface for the knowledge sharing portal developed for the arts/cultural organisations. However, the U&I community required slightly more functionality and it is the knowledge sharing portal developed for them that will be the focus of this section. Once logged into the system a user has several options available to them concurrently, see Figure 4. Essentially the screen is broken up into two areas of navigation, the top navigation bar and the categories. The top navigation bar is comprised of: My Files, Upload, Preferences, Categories, Micro Cloud, Last Upload, Search and Search History.

Figure 3. A screenshot of the knowledge sharing portal interface.

The My Files section comprises a record of existing files that have been uploaded by a user and a number of options are available, arranged horizontally, with each file dependant upon its type. All files can have supplementary files added to them here. The file type is displayed here by its extension, for example .txt or .jpg. The title and first 30 characters of description are displayed; these are both hyperlinked to the actual file. Each file also has the capacity to have its category and sub-category to be re-defined. Each file also has the capacity here to be viewed or listened to and also to be downloaded. The download of text files with the extension of .txt are converted automatically to the Adobe PDF file format. There is a caveat on files that need a specific browser plug-in, for example, files that might use the free Quicktime plug-in to be rendered by a browser are usually not downloadable, this can only be done if a user is in possession of the paid for Quicktime PRO plug-in. The next option available on all file types is the capacity for it to be removed from the knowledge sharing portal. If this option is selected then a confirmation dialogue box is displayed which allows the user to accept or reject the file removal operation. The next option is available only to those file types that can be microtagged or intra-tagged. This option basically allows the user to edit or remove any existing intra-document tags that have been created or add new ones if necessary. The final option on this page enables a user to toggle between allowing others to add tags to their documents or not. The next option on the top navigation bar is for a user to upload files with. The initial option uses a Microsoft Windows style browse interface, the user is presented with the option to browse around their computer/network for a given file that they wish to upload. Once a file has been uploaded the user is required to give the file a title and description as well as allocate a category and relevant sub-category. The user can elect at this point to enable other users to tag the file before saving their choices. The next screen that is presented is typically the document ready to be intra-tagged, if the file type is appropriate. The user can for example select an area of an image by creating a marquee over the appropriate area and then ascribing a tag to that marquee area by using the Add Tag button. With text, the process is similar, text needs to be highlighted/selected and then a tag can be ascribed using the Add Tag button towards the bottom of the screen. Irrespective of whether the file is text or an image, the user can define as many tags as they wish.

47

ISBN: 978-972-8924-82-9 © 2009 IADIS

Further, selected areas to be tagged can overlap. The user is also given the option to select tags for deletion as well as adding more files to be linked to the document. This latter option takes the user back to the upload process already described.

Figure 4. The user menu available to the U&I community.

The preferences function is the name of the next option available to the user on the top navigation bar. Once selected the user is given choices as to how they would like text and tags to be rendered by the browser. The choice of preference includes text size, foreground and background colour of tags in a document as well as foreground and background colour of text in a document. The Categories and Micro Cloud are the next two options available and have already been described. The Last Upload section supplies a list of the latest uploaded files by the entire community. The latest uploaded files are presented in descending chronological order with the latest file displayed at the top. Other details shown here for each file include: the type of file, the username of the uploader, the date the file was uploaded, the number of tags attached to the file, an indication of whether or not the file can be intra-tagged as well as the ability to do so and finally a way to navigate to the file. The Search and Search History options enable a given user of the knowledge sharing portal to search for content based on a given keyword or phrase. The search is by default enabled to search tags through out all fields. However, if required, a search can be narrowed by filtering which fields to conduct the search on. The available fields were searches can be filtered through, are by the uploader, by title, by description, by category, by sub-category and by micro tag. The search history option enables a user to view a list of their most recent searches within the knowledge sharing portal. This functionality enables a more streamlined workflow, the concept being that a user can view and activate their most recent searches rather than have to conduct the search again.

5. EVALUATION RESULTS AND DISCUSSION Testing of the knowledge sharing portal was an essential process and was undertaken by specific discrete groups. Three groups assessing specific areas carried out the testing, arts professionals, software testers and non-regular users. Arts professionals assessed the general functionality of the knowledge sharing portal by logging in to the system and carrying out specific tasks, such as: testing the upload feature, testing the tagging features, testing the tag cloud and search features, assessing whether the knowledge sharing portal facilitates an impact value towards their work. Software testers tested specific functionality of the knowledge sharing portal for errors, again by logging into the system and carrying out pre-determined tasks in a similar fashion as the arts professionals. Finally, non-regular users looked specifically at content and usability and testing as the previous two groups. User testing required a number of user roles to be looked at such as expert or non-expert users, additionally a user guide was created to be used with worksheets for conducting the testing and was carried out on the 30th of May 2008. Functional testing had been carried out as an ongoing process in-house during the course of development, however, user testing was an eminently sensible process to undertake to investigate the usability and functionality of the knowledge sharing portal development to date. A major factor here was to get feedback and opinions from people who had not been involved in the development of the project and were viewing the site for the first time, see table 1 for the types of issues identified.

48


Testers represented three domains: expert specialists from the arts professionals domain, technical experts from the computing domain and non-regular users. Testers volunteered from the Ambition project team, PhD researchers from the Institute of Applicable Computing as well as administrative staff and Post Graduates from the Visual Arts/Media field at the University of Bedfordshire. Testing was carried out at the University of Bedfordshire, except that done by the Ambition project team which was conducted off-site. All testers were given essentially the same tasks to conduct with the aim of: Logging in to the system, testing the knowledge sharing portal upload, testing the knowledge sharing portal tagging, testing the knowledge sharing portal tag cloud and search, and assessing whether the knowledge sharing portal facilitates or helps the user with their work. Testing was carried out on the PC and Macintosh platforms with both Internet Explorer and Firefox browsers tested on the PC and Firefox on the Macintosh. Both Mac OSX and Windows XP were used during the course of this testing. Table 1. Primary issues identified for correction. No. 1 2 3 4 5 6 7 8 9 10

Error Description Confusion with Homepage navigation Confusion with knowledge sharing portal navigation Confusion with site terminology Failure to save tags/Site failure Issue/boredom with repetition of tagging Difficulty in Macro Tagging Difficulty in Micro Tagging Uncertainty about capitalisation of tags Confusion with Search facility Difficulty with tag browsing

Occurrences (out of 10) 3 3 5 4 2 1 4 2 2 5

On the 27th of November 2008 the knowledge sharing portal was presented to fifteen U&I community members during an online conference using the Elluminate online collaboration tool. The audience were given a presentation as to how the knowledge sharing portal operated and the types of functionality it had. This was followed by a brief exercise in using the knowledge sharing portal that the audience participated in. After the exercise, participants reflected on their experience during the exercise and sought some further clarification on either functionality or scope of the knowledge sharing portal. Members of the audience expressed that there was an amount of similarity with regards to the intratagging principle being offered by the knowledge sharing portal and by the online service offered by Diigo (www.diigo.com). One essential point came up during this discussion and that was the issue of rights access and who should have it. Basically there were two schools of thought on the issue, one was that only selected members from the U&I support team should have rights to upload and tag material, while the other line of thinking was that everyone in the U&I community should be able to upload and tag material. The case for only selected members being able to upload and tag was essentially that a certain level of quality of service could thereby be maintained. The selected team would be instructed as to aesthetic content parameters and could ensure that these were adhered to when the wider community fed material to them. The alternative viewpoint was that the whole community should have rights to upload and tag material as they saw fit. A uniform quality of service could not be as tightly managed but peer support/encouragement could help ensure that an appropriate quality of service was produced by the community as a whole. Subsequently (January12-23 2009), the U&I community were offered a questionnaire that was actioned by Glenaffric, an independent consultancy brought in as evaluators on the U&I project. Five questions from the survey were directly concerned with the knowledge sharing portal and its usage. The following is a summary of the relevant survey findings. From the twenty responses, 35-55% of them would be likely to contribute to the knowledge sharing portal, the most likely file type to be uploaded would be textual (42%), that uploading and tagging files was seen as the most important functionality for this kind of repository, that 35-45% of respondents would refer to it during the course of their current projects and finally that the majority of respondents (60%) would be most likely to use the knowledge sharing portal as a resource in future projects. Therefore, we can conclude that U&I community perceived the knowledge sharing portal as beneficial not just during this project lifecycle but quiet likely in future projects as well.

49

ISBN: 978-972-8924-82-9 © 2009 IADIS

6. CONCLUSION AND FURTHER WORK The basic premise to future proofing the knowledge sharing portal is through the addition of an XML conversion module. By converting files to XML before or during the upload step intra tagging of more filetypes becomes possible, see Figure 5. TXT

PDF

XML Converter Module

Tagged XML Document

WORD

Figure 5. Improved knowledge sharing portal concept.

By using XML based files on the knowledge sharing portal the system can accept numerous file-types for various and multiple purposes. Additionally the use of XML can facilitate the use of embedded URLs from within a given tagged document. For example, by pulling in XML data for the knowledge sharing portal, the system should be able to produce facilitation for a number of applications serving differing needs. Facilitation for applications could range across a datasite, a mashup, or a data integrator to name but a few. The development of the knowledge sharing portal has been described, the portal acts as a means to allow two dispersed user communities to publish and search for relevant information, including audiovisual material, through a Joomla mediated web interface. During this development, a rapid application development approach has been used throughout. The approach was an iterative process of building and testing with lessons learned or observations made from the testing phase being implemented in the next building phase. The system was previewed by members of the user communities with further changes and enhancements being made to the system before it was trailed/critiqued by the wider user communities for further development. Subsequently, the U&I community were offered a questionnaire that was actioned by Glenaffric, an independent consultancy brought in to evaluate the entire U&I project. Results from this survey suggests that the user community perceive the knowledge sharing portal as beneficial not just during this project lifecycle but quiet likely in future projects as well.

REFERENCES Bessis, N., Gaitan, A., Shukla, M. and Stephens, J. (2008). Using Next Generation Service Oriented Technologies to Assess Cultural Handling of Information Communication Technology, International Conference on Computer Science and Information Systems, Web Based Communities 2008, IADIS, 24th-26th July, Amsterdam. Glaser, B.G. & Strauss, A. L. (1967). The Discovery of Grounded Theory. Strategies for qualitative research. Chicago: Aldine. Hassan, Yusef and Monteroa Víctor Herrero-Solana (2006), Improving Tag-Clouds as Visual Information Retrieval Interfaces, International Conference on Multidisciplinary Information Sciences and Technologies, InSciT2006. Mérida, Spain. Hildrech, P., Kimble, C., Wright, P. (2000), "Communities of practice in the distributed international environment", Journal of Knowledge Management, Vol. 4 No.1 Mathes, A. Folksonomies – Cooperative Classification and Communication Through Shared Metadata. Computer Mediated Communication - LIS590CMC, December, 2004.

50


THE USE OF AN ONLINE HEALTH COMMUNITY ON OVERWEIGHT: A MEMBER PERSPECTIVE Åsa Smedberg The Dep. of Computer and Systems Sciences, Stockholm University/KTH Forum 100, S-164 40 Kista, Sweden

ABSTRACT Today, there are lots of open web based communities for people with different mental and physical health issues. Among these are health communities for people who suffer from established bad habits, who try to lose weight or quit smoking, for example. These communities aim at helping people change habits, and thereby reduce the risk of getting severe illnesses. The community members discuss their concerns about everyday situations, and they post requests about treatments, exercises, motivation, etc. The question is how these communities manage to trigger learning and behavioral change among the members. In this paper, results from a study of the community members’ own experiences are presented. The results are based on a questionnaire to the members of a web based community on overweight present on the Swedish Net Doctor site. The results showed how the members found community participation to be beneficial, but they also pointed out some difficulties related to their learning processes. KEYWORDS Health communities; conversations; preventive healthcare; bad habits; learning.

1. INTRODUCTION People who like to discuss health-related issues and to search for health information constitute a large and even increasing group on the Internet (Fox and Fallows, 2003), and patients as well as citizens are getting more and more empowered due to new ways of using the Internet (Fox et al., 2005; Korp, 2006). Studies give that web based communities on health issues are used on a daily basis by people with different physical and mental conditions. Through web based health communities, people can share experiences, and offer mental support and advice on how to cope with different health conditions (Maloney-Krichmar and Preece, 2005). The importance of these web based systems is also manifested in the development of national portals for health services, such as NHS Direct Online and the Net Doctor sites. In general, these national portals facilitate access to health information sources, communication with medical-trained people and group communication with other patients and citizens who offer their personal perspectives, for example. A goal is to make available appropriate, and even proactive, health management (Milicevic et al., 2005). Studies show that patients who interact online with other patients benefit from this (Walther et al., 2005), and chronically ill patients manage to cope with their stressful life situations better (Josefsson, 2007). The new web based systems bring with them that patients can act both as consumers and producers of medical information. Different coping activities on the Internet include information seeking, seeking of social support and to support others, and they allow patients to get strengthened in their position to challenge the medical expertise. Through online communities, both strong and weak-ties relationships are possible to maintain (Haythornthwaite, 2006). The wide-spread usage of Internet and the new Internet services have created a lot of new possibilities. However, in what ways they affect our health or well-being, and our social and work life, are not given or deterministic; instead, consequences are highly dependent on context (Kling, 2000). It is not that new Internet services automatically lead to the public being better informed, or that more available health information leads to a certain practical consequence. For example, people may be seeking information on the Internet in order to find a doctor they can feel confident about. But they can also look for alternatives to doctor-mediated medical care, for example, information about health, herbal medicine or postoperative care. The use of

51

ISBN: 978-972-8924-82-9 © 2009 IADIS

medical sources on the Internet can help people receive “...alternative medical advice or information about issues that their doctors don’t address well.” (Kling, 2000, p. 226) While surgeons may be experts in the domain of surgery, they may not be as good at explaining the recovery procedures to the patients, for example. Such a situation makes room for additional information to be shared on the Internet, and for other knowledgeable people to step forward. Besides the web based health communities on diseases of different kinds, there are also those that focus on bad habits, or unhealthy behaviors, such as eating problems, smoking, and alcohol and drug problems (Smedberg, 2004; Smedberg, 2008.a-b). Especially bad eating habits and overweight or obesity has become a problem to many people (Timperio and Crawford, 2004; Stevens et al., 2008). These life-style problems can lead to critical health conditions, and changing the habits is important in order to reduce the risk of getting illnesses such as heart conditions, diabetes and asthma, for example. Preventive health care is therefore recognized as important in order to help these groups of people from becoming patients of the health care system. However, a problem for the health care is that treatment programs tend to have a short-term effect; people who suffer from overweight often regain weight after finishing the treatment programs, for example (Byrne et al., 2004). Instead, ongoing social support is a crucial factor for developing a life-style that lasts (Baughman et al., 2003; Norman et al., 2007). Weight management behaviors can be supported by offering choice, providing feedback and stressing the importance of self-management (Stevens et al., 2008). It is in this context that the national health sites can play an important role; through advice from experts and through ongoing conversations in the web based health communities, long-term improvement of people’s health is possible to achieve. Senge (2004) emphasizes the value of making distinctions between single events, patterns of behavior and the systemic structure. He explains single events as the first level of reality to which we can only react, the patterns of behavior as the second level that we can learn about through trend analyses and long-term historical changes, and then respond to, and the systemic structure as the last level of reality that addresses the question of “What causes the patterns of behavior?” The last level involves generative learning, something that requires that we question our assumptions, change our mental models and restructure our views of reality. This last level can thus help us stop a behavior or habit from being repeated in the future. What motivates and triggers a certain behavior is then necessary to understand. We can help each other by explaining, criticizing, sharing and motivating, things that become important parts of collaborative learning (Soller and Lesgold, 2000), and opposing ideas are also useful for reflection and learning (Baker et al., 2002). Changing behavior can be difficult to succeed with. There are several theories regarding elements to measure. However, some common themes can be seen when the different theories are compared (Noar and Zimmerman, 2004). Examples of elements to measure are: attitudinal beliefs (appraisal of the positive and negative aspects of behavior and expected outcomes of the behavior), self-efficacy (belief in one´s ability to perform the behavior), normative and norm-related beliefs and activities (belief that others want you to engage in the behavior, that others are engaging in the same behavior, it may also include support of others), risk-related beliefs and emotional responses (belief that one is at risk if one does not engage in the behavior, and a belief that the consequences may be severe) and intention (intend to perform the behavior, setting goals or making a commitment to perform the behavior). Even though there is a common understanding that web based communities are helpful for groups of people who suffer from different health conditions and who need support in order to cope and manage, there is a knowledge gap regarding characteristics, needs and challenges of the community members, and what kind of conversations they appreciate and value in their struggle for a healthier life-style. This ‘behind the scenes’ knowledge will be addressed in this paper.

2. THE STUDY The study presented in this paper is based on a survey with members of a web based overweight community.

2.1 The Overweight Community and its Members The web based community and its members targeted for the study were found on the Swedish Net Doctor site. The community aimed at helping overweight and obese people to come together to support each other in

52


their struggle to lose weight. Anyone could become a member of this community, and thereby start to post requests and answers in the conversations. The users could use alias names if they wanted to. They could also give a description of themselves and their interests, and also add a picture of themselves.

2.2 Method The community members’ experiences of using the Swedish Net Doctor community on overweight were investigated through an online questionnaire. The questionnaire was divided into questions on background information (10 questions) and questions related to experiences of using the Net Doctor community (27 questions). There was also a final question about contacting the respondent for an eventual follow-up. The survey focused on the members’ experiences from community activities, but it was also designed as to mirror some of the important elements related to behavioral change. The questions were constructed in different ways; 19 of them were multiple choice questions, ranking and boolean questions, with the possibility to add comments, and further four questions asked for background facts, such as gender, age and weight, e.g., while 15 questions were open questions that asked for personal reflections on different matters related to the members’ use of the web based community. The community members were asked to participate in the survey through a blog post on the Swedish Net Doctor site. The number of community members who participated in the survey was 30 in total. The results from the survey were analyzed foremost from a qualitative perspective, but some quantitative analyses were also made.

2.3 Results This section presents the results of the survey. In order to gain some contextual knowledge regarding the survey participants, we will start by presenting some background information that the participants shared through the survey. After that, the participants’ experiences and reflections on their own habits as well as the community activities are presented. Their usage of the community, their opinions about the issues that were raised in the conversations and the way the conversations were carried out were things that they reflected upon. Also, their own progress as well as things they found challenging will be addressed in this section.

2.3.1 General Characteristics of the Community Members in the Study Almost all of the community members that took part in the study were women, except for two men. The age spanned from 20 to 65, and the average age was 36. Most of the overweight members in the study had participated in the community more than 12 weeks (75%), while the remaining members had participated from less than a week up to six weeks. Almost 40% of the members visited the community several times per day, and when those who visited it once a day were included, the number increased to 57%.

2.3.2 Established Bad Habits – Changes and Challenges The participants from the overweight community had a Body Mass Index (BMI) value of 31 in average. About 28% of the participants had a BMI between 25 and 29, which is considered overweight, 48% had a BMI of 30 and higher, which is the weight range used to define obesity, and the rest, 24%, had a BMI between 20 and 24 which is normal weight1. The highest BMI was 54, and it belonged to a woman who had been a member of the community for less than three weeks. In average, the members had lost 4.6 kilos each during the time they had been a member of the community. The member who had lost most weight was a woman with a present BMI of 29. She had lost 23 kilos in total during her time as a member (more than 12 weeks). There was also one member who had gained eight kilos during her time in the community (more than 12 weeks). When trying to change habits, it is important to acknowledge the importance of change. Among the members from the overweight community, as many as 70% saw impending risks with continuing with unhealthy habits. In their answers to what risks they saw, they gave examples of different health conditions such as diabetes, heart conditions, high blood pressure, fatigue, knee pain and early death. About 64% of the 1 The weight ranges are based on recommendations by health centers, such as Centers for Disease Control and Prevention, USA, and applied in health sciences, for example Beuther and Sutherland (2007).

53

ISBN: 978-972-8924-82-9 © 2009 IADIS

members from the overweight community said that their habits had great impact over the way they felt. Another element of importance when trying to change behavior is self-efficacy, or believed ability to change. The community members were therefore asked to rank their ability on a scale from 1 (low ability) to 10 (great ability). The average number for all participating members was approximately 8.4. The lowest ranking came from a member from the overweight community who ranked her ability to 1, i.e., the lowest level of ability. Her BMI was 54 and she declared no weight change during her time in the Net Doctor community, but she had only been there for less than three weeks. She expressed a certain feeling of despair by saying that: “[…] I cannot take it any more”. On the other side of the scale, those who ranked their ability as 10 had in average a BMI of 29. It is always difficult to determine what makes people change behavior, what triggers a change. Since people and their environments are complex systems, there are no clear relations between different elements. However, there were examples of established bad habits that the community members had changed since they became members of the community. Some of the examples given by the members of the overweight community are listed below (as quotations): - I have smaller meals - I exercise more regularly - I have found new ways to exercise - I have gained new insights about my problems and I am more aware of what I eat now - I go for a walk more often and engage more in daily activities - I do not “over-eat” so often anymore as a result of emotions About different habits that the members experienced as being especially difficult to change, some examples can be seen below (as quotations): - Old weekend habits with wine and food - To stop the addiction of sugar has been the most difficult for me, and also for others in the discussion forum - My habit of eating candy - I have quite a big stomach and can therefore eat a lot without being full - To go shopping is hard, with all temptations - The food. To choose the right thing when you feel for something unhealthy - To stop drinking alcohol - To have exercises every day. I am still a “sofa potato” About the way the community conversations had effect on their well-being, 57% said that they felt that both their physical and mental/social well-being had improved since they joined the community, 24% experienced either a physical or mental/social improvement, while 19% did not experience any difference. No one said that their well-being had decreased.

2.3.3 Experiences from the Community Conversations In the answers to the question about what characterizes a good online discussion, the following were said: - That you respect each others’ opinions - Different points of views, a nice tone and no lecturing - That no one is rude or openly critical towards the others - Openness - Room for different views but without attacks - Different opinions are presented in a well-meant way - Good discussions take place when curious, open and humble people participate. It goes for both real-life and on the Net… - That you “listen” to each other - A give and take situation - You share your own experiences and help each other The community members presented also some general reflections, some pros and cons, of using the overweight community. From one of the community members the following statement was given: “The best thing is all the members who give their support and care!”, and “A good community spirit” was another example of a positive thing with the community. About the cons with the overweight community, one member explained that: ”Sometimes it feels like people are experts and patronizing, they tell you what diet that is the best, etc.” Another comment from a member was: “There are fools out there who give lousy advice

54


and who participate only to market different medicine, bad diets and websites.” Another member put forward a risk: “A risk is that it can be too anonymous, you know the others but you still don’t in a way. Others can tell you lies, you cannot tell whether people lie or tell the truth. But you may assume that they do.” Regarding what kind of discussion issues that the community members appreciated the most, table 1 shows the way the members ranked five categories of issues. Table 1. First choice regarding preferred type of discussion. Type of discussion issue Fact queries

Occurrence 52%

Questions on incentives

22%

Questions on setbacks

0%

Questions on obstacles

13%

Statements

13%

When the members were asked to motivate their first choice of discussion, different ways to benefit from the discussions were shown. “It is important to get new angles of approach and new research results.” was one comment from a member of the overweight community who ranked fact queries the most. Another remark from someone else who favored the fact queries was: “It is foremost these [fact queries] that I choose since it is my field of interest. I am also a food counselor and therefore I like to share my knowledge in these discussions.” Other remarks from the overweight members who preferred fact queries were: “You learn how to eat healthy food.” and “I use it to reflect.” About the second most preferred type of issue in the overweight community, the questions on incentives, the following were said: “I get pepped up when I am feeling down.” and “I collect a lot of tips and so much energy that I can keep up with this change.” The community members were also asked to reflect upon different aspects of existing community conversations. The members were supposed to choose whether a specific aspect should increase, decrease or stay the same. On next page, figure 1 shows the results from this question. According to the results, the three elements that some of the members thought could decrease in number were “answers with deviating positions”, “answers that include good advice” and “competitions”. Regarding the first two elements, the percentage was quite low (5% each), while the last one, about competitions, had a percentage of 29%. An interesting result is the discrepancy between the members’ view of “answers with deviating positions” and “answers offering new perspectives”. While deviating positions were recognized by some of the members as replaceable, and 57% preferred these to stay the same, 62% regarded new perspectives as something worth increasing. What can also be seen from the results is that “questions on incentives” were recognized as something that the community would benefit from having more of; more members liked to increase the number of discussions on incentives than those who wanted this type of element to stay the same (62% vs. 38%). The community members also reflected upon their impressions of the ways the other community members expressed different positions. Most of the members thought that the different positions were presented in an empathetic way (57%), while 29% of the members believed that this kind of postings most often were presented in a neutral way. Only 14% thought that they most often were expressed in an unfriendly way.

55

ISBN: 978-972-8924-82-9 © 2009 IADIS

100% 90% 80% 70% 60% 50% 40% 30%

The same

20%

Decrease

10%

Increase

0%

Figure 1. The need to increase, decrease or to let the community elements stay the same.

2.3.4 Different Roles - Active Participation vs. to Follow the Discussions About half of the members in the study (52%) had at least once started a discussion in the community. “When I want to share something fun or when I am feeling bad.” was a comment showing two situations when discussion issues were posted. From one who had never posted an issue, the following was said: “I read most of the time, but I do not fear to write if I consider myself to be able to contribute to the discussion.” Another comment was: “I participate actively when I have the time, if not, I just go through the discussions briefly and interfere in the ones that concern me the most.” However, only less than 5% of the members said that they had questions that they did not dare or want to post. A comment from a member who had hesitated to post a question was: “I guess everyone has questions they are embarrassed to ask.”

3. CONCLUSION In this paper we have addressed the community members’ own experiences of participating in a web based community on overweight. The purpose was to learn more about the community members behind the online conversations. Their views on different aspects of community conversations and their experiences of changes in weights and life-styles were investigated. The results gave that the community members experienced positive effects on their well-being due to their community participation. Also, almost all of the community members in the study experienced that even when different positions were posted by other members, the ideas were presented in an empathetic or neutral way. This indicates that the community was overall regarded as friendly.

56


What was also concluded from the study was that the fact queries were the most appreciated ones of the posted questions, followed by questions on incentives. At the same time, the members expressed that they were quite satisfied with the present number of fact queries, while they wanted to see more of conversations on incentives, personal experiences and new perspectives. The results showed examples of improved eating and exercise habits among the community members. However, to determine the effects actually caused by the fact that they participated in the community is not possible from the study. The main purpose was to look for examples of experienced life-style changes that the community members wanted to share. At the same time, there were also examples of situations in their everyday lives when they thought it was hard to choose the healthy alternative. This reflects the challenging work of trying to change behavior. The methodological approach used for this study could most likely be applied to other web community settings in the field of life-style problems, such as smoking, alcohol and drug abuse. Since one of the purposes was to search for experienced outcomes and changes in life-styles, other web communities of this kind could benefit from having similar community member studies made. Also, questions that map out what the community members prefer and not can be helpful when designing new web based communities.

ACKNOWLEDGEMENT I want to thank the members of the Net Doctor community on overweight who allocated their time to answer the survey questions. This paper is based on their participation: their reflections and ideas. I would also like to thank two colleagues at my department, Gunnar Wettergren and Sven Olofsson, who assisted me in preparing the technical survey application.

REFERENCES Baker, M. et al., 2002. Designing a Computer-supported Collaborative Learning Situation for Broadening and Deepening Understanding of the Space of Debate. Proceedings of the Fifth International Conference of the International Society for the Study of Argumentation. Amsterdam, Netherlands. Baughman, K. et al., 2003. Biopsychosocial Characteristics of Overweight and Obese Primary Care Patients: Do Psychosocial and Behavior Factors Mediate Sociodemographic Effects? Preventive Medicine, No. 37, pp. 129-137, Academic Press. Beuther, D. A. and Sutherland, E. R., 2007. Overweight, Obesity, and Incident Asthma - A Meta-analysis of Prospective Epidemiologic Studies. Am J Respir Crit Care Med, Vol 175. pp 661–666. Byrne, S. et al., 2004. Psychological Predictors of Weight Regain in Obesity. Behaviour Research and Therapy, No. 42, pp. 1341-1356. Centers for Disease Control and Prevention, Dep. of Health and Human Resources, USA: Defining Overweight and Obesity. Web site accessed on the 20th of January 2009, http://www.cdc.gov/nccdphp/dnpa/obesity/defining.htm. Fox, N. J. et al., 2005. The ‘Expert Patient’: Empowerment or Medical Dominance? The Case of Weight Loss, Pharmaceutical Drugs and the Internet. Social Science & Medicine, 60(6), pp. 1299-1309. Fox, S. and Fallows, D., 2003. Internet Health Resources: Health Searches and Email Have Become More Commonplace, but there is Room for Improvement in Searches and Overall Internet Cccess. Pew Internet & American Life Project, 16 July 2003. Haythornthwaite, C., 2006. The Social Informatics of Elearning. Information, Communication & Society (ICS) 10th Anniversary International Symposium. York, England. Josefsson, U., 2007. Coping Online – Patients’ Use of the Internet. Doctoral thesis, Report 37, Dep. of Applied Information Technology, IT-University of Göteborg, Sweden, ISBN 978-91-628-7080-5. Kling, R., 2000. Learning about Information Technologies and Social Change: The Contribution of Social Informatics. The Information Society, No. 16, pp. 217-232. Korp, P., 2006. Health on the Internet: Implications for Health Promotion. Health Education Research – Theory & Practice, 21(1), pp. 78-86. Maloney-Krichmar, D. and Preece, J., 2005. A Multilevel Analysis of Sociability, Usability and Community Dynamics in an Online Health Community. ACM Transactions on Computer-Human Interaction, 12(2), pp. 201-232.

57

ISBN: 978-972-8924-82-9 © 2009 IADIS

Milicevic, I. et al., 2005. Making Progress towards User-orientation in Online Public Service Provision in Europe. Proceedings of eChallenges IST 2005 Conference. Ljubljana, In Innovation and the Knowledge Economy, IOS Press. Noar, S. M. and Zimmerman, R. S., 2004. Health Behavior Theory and Cumulative Knowledge Regarding Health Behaviors: Are We Moving in the Right Direction? Health Education Research, 20(3), pp. 275-290. Norman, G. J. et al., 2007. A Review of eHealth Interventions for Physical Activity and Dietary Behavior Change. Am J Prev Med., 33(4), pp. 336-345. Senge, P. M., 2004. The Leader’s New Work: Building Learning Organizations. In Starkey, K., Tempest, S. and McKinlay, A. (Eds.) How Organizations Learn (pp. 462-486), 2nd edition, Thomson Learning. Smedberg, Å., 2004. Learning through Online Communities: A Study of Health-Care Sites in Europe. Proceedings of eChallenges Conference. Vienna, Austria. P. Cunningham and M. Cunningham, Eds. IOS Press, pp. 1333-1339. Smedberg, Å., 2008.a. Learning Conversations for People with Established Bad Habits: A Study of Four HealthCommunities. Int. J. Healthcare Technology and Management, 9(2), pp. 143-154. Smedberg, Å., 2008.b. Online Communities and Learning for Health - The Use of Online Health Communities and Online Expertise for People with Established Bad Habits. Doctoral thesis, The Dep. of Computer and Systems Sciences, Stockholm University, Sweden, ISBN 978-91-7155-689-9. Soller, A. and Lesgold, A., 2000. Modeling the Process of Collaborative Learning. Proceedings of the International Workshop on New Technologies in Collaborative Learning. Awaji-Yumebutai, Japan. Stevens, V. J. et al., 2008. Design and Implementation of an Interactive Website to Support Long-Term Maintenance of Weight Loss. Journal of Medical Internet Research, 10(1): e1. Timperio, A. and Crawford, D.A., 2004. Public Definitions of Success in Weight Management. Nutrition & Dietetics, No. 61, pp. 215-220. Walther, J. B. et al., 2005. Attributes of Interactive Online Health Information Systems. Journal of Medical Internet Research, 7(3): e33.

58


THE NEED FOR TWO COMMUNITIES John S. Murnane ICT in Education and Research The University of Melbourne PARKVILLE, 3010, Melbourne, Australia

ABSTRACT The onset of age-related problems can be delayed by activity and mental stimulation, and for the aged, the Internet can supply several elements scarce or missing from their lives. The Web, and particularly eMail, can provide by far the best link to family, friends and the World, and is a powerful and interesting environment which can be indulged in without the need for organised group activities. Experience with a small research project at the Old Colonists’ Association of Victoria shows over 80’s in supported Hostel accommodation with no experience of computers can learn to use eMail and the Web independently, but the learning process is slow and teacher-intensive, and this can make it expensive. Two different communities need to develop if the project is to become self-sustaining: Hostel residents who can use the Internet need to take over responsibility for teaching new retirees, and family members have to understand the importance and necessity of maintaining regular communication to keep the residents’ interest. In effect, family and residents need to see themselves as a community. KEYWORDS Aging, email, education, peer mentoring, family.

1. INTRODUCTION For many years it has been almost accepted wisdom to regard computing as a young person’s world. Morris and Ballard (2003, p. 169) state that “although the research on computer anxiety and attitudes with older adults has yielded mixed results, older adults experience more anxiety and more negative attitudes toward computers than younger adults.” In fact in some quarters, it has almost become the fashion to disparage the ability of anyone not born into ‘the computer age’ to take up, or even see the utility of the computer and its inherent communication possibilities. Prensky (2005, p. 29) refers disparagingly to anyone too old to be in school in 2005 as a “digital immigrant.” Butler (2005, p. 7) considers everyone born later than 1982 has “an intimate” relationship with the Internet, implying those born before this date do not. TV and radio talk-back hosts sometimes express the opinion that older people cannot use communication technology, or only use it with difficulty. As late as March 2006, Mark Colvin, host of ‘PM,’ the Australian Broadcasting Corporation’s flagship current affairs program, remarked that he “still can’t program [his] VCR” (Colvin 2006). This was apparently meant as a light, ‘throw away’ remark, but it and similar utterances do nothing to encourage the non-computer literate to try and use the technology, in fact they build a culture of fear amongst older people that they would not be able to cope. This attitude is very unfortunate because the Internet, in its World-Wide-Web form, might have been especially designed for the elderly or disabled. Potentially liberating in ways even surpassing its importance for the young and mobile, it can provide a dimension otherwise totally lacking in the life of someone limited in their mobility or confined by circumstances to a single building. In a trial providing Internet training to retirees, White et el. (2002, p. 213) found “most elderly participants … learned to use the Internet and the majority continued to use it on a weekly basis.” Morris and Ballard (2003, p. 169) note that “exposure to computers and training in basic computer skills may result in more positive attitudes, greater motivation to learn, and reduced anxiety levels.” (See also White et el. (1999).) Leaving aside medications, Watari and Gatz (2002, p. 34) list the protective factors “with the most convincing support” against Alzheimer’s disease as “education and mental stimulation.” In fact Seniors

59

ISBN: 978-972-8924-82-9 © 2009 IADIS

health in general is closely tied to their mental stimulation (Hanson, et al. 2007, Leonard 1993, Rönnberg 1998) and an Internet connection to the World, along with other applications routinely provided by a computer can make a very positive contribution to this. Rönnberg (1998: 393) asserts that “only rarely do nursing homes meet all the psychological needs of their residents. The environment has been described as one of limited privacy and isolation from the outside world.” Little appears to have changed since 1998. Regular contact with relatives and friends outside the retirement home is another area often mentioned in connection with the promotion of Senior’s health (Downs, et al. 2006). Dobbs et el. (2005, p. 84) found that “increased resident activity participation was associated with two measures of family involvement: the amount of time the family reports being socially engaged with the resident and the family's degree of involvement in assessing resident preferences.” Much of the government-generated and sponsored material to which Seniors require access is most easily found electronically (Tran and McComb 2004). Indeed governments seem particularly keen to advance this type of ‘one-stop-shop’ communication. The Information Victoria Website, which includes Aged Care in Victoria, aim is “to provide quick and easy access on all State Government information and services to the Victorian community” (Information Victoria 2002). A search of the families.gov.au website provides a wealth of information about retiring, as does www.seniors.gov.au. At the same time there appears to be an assumption that once safely inside a retirement home, responsibility for monitoring and acting on such material can be safely passed to the home Management, or relatives, since access to it is usually unavailable. Although many people have to be cared for by full time professionals and mentally are no longer up to the task of running their own financial and legal affairs, this is not the case for all, and someone who is capable of looking after their own dealings, or at least having an intelligent say in them, should not have access to this form of information denied by default. “Loss of control and lack of autonomy lead to learned helplessness” (Rönnberg 1998, p. 393). The profusion of ‘seniors’ websites, including an excellent one by the Australian Government as part of www.seniors.gov.au, shows that older people can use computers and the Internet. Nevertheless, the proportion of Australian retirement homes with Internet access is small. Australian Government and AusStats Webpages provide no information on Internet access in retirement homes at all. Research in 2005 at Sydney University recognised “that very little is known about older people and their internet use” (Older people and information technology 2005, p. 1). Yet what research there is on the benefits of being able to use internet facilities suggests they should be mandatory in all accommodation for the restricted or disabled (Hancock, et al. 2006, White, et al. 2002). As well as benefits derived from more frequent contact with relatives, retirees participation in the general environment of the Internet and the ability to ‘go anywhere in the world,’ is also a strong attraction: using the Web is a pre-eminent leisure activity. Scarmeas et el. (2001, p. 2236) found that “the data suggest that engagement in leisure activities may reduce the risk of incident dementia.” Yet in Australia, retirement complexes rarely provide Internet access for their residents. Usually only those who have the experience, their own equipment and can pay for an ISP can participate. Given the advantages of an Internet connection this is very unfortunate. The remainder of this paper considers teaching, community and family issues associated with retirees learning to use the Internet, based on a small research project at the Old Colonists’ Association of Victoria (Murnane 2007, 2008). Experience has shown that the teaching methodology used, which was essentially determined by the retirees themselves, has deterministic and vital implications for the way the online community is developing, both within the Hostel and on the Web.

2. RESEARCH SETTING The research on which this paper is based on was carried out at the Old Colonists’ Association of Victoria Leith Park low care Hostel in Melbourne, Australia. Leith Park consists of an extensive area with 118 independent living cottages, a low-care Hostel with 52 residents, including a 12 room dementia wing, and is in the process of building a high-care Nursing Home. To make the project attractive to Management, the ICT in Education and Research Cluster from the University of Melbourne agreed to supply two computers linked to the Internet, arrange and finance the Internet Service Provider (ISP) and provide training for the residents for 12 months. At the end of this time the equipment would become the property of Leith Park. In return

60


some staff and residents agreed to the collection of data. This involved short questionnaires for residents and staff, but came predominantly from participant observation with residents learning and using the machines. Residents were offered instruction in computing and electronic communication skills, initially the use of eMail. Due to the small research budget, this instruction was limited to an average of three hours a week. All residents, whether they joined the ‘research’ or not, were free to use the computers and take advantage of the instruction provided. The project began in August 2007 has been running for over 18 months. Due to its very limited budget, the teacher is also the researcher. This restricts the number of residents involved in the teaching program, but results suggest that they can be regarded as typical for retirement complexes across Victoria

3. THE ‘STUDENTS’ Most retirees enter the Hostel over the age of 80. Those in our teaching program have numbered nineteen: seventeen female, two male, ages 84 to 97. It is in the nature of this environment that new residents will come and others will move on. We began with a group of six, and two of these no longer participate. Additionally, three retirees from the Independent Living Units, ages 66 to 87, all of whom have their own computers, come occasionally for lessons in specific applications.

4. TEACHING Eight residents, hereafter the students, had never used a computer before, including five who had never used a keyboard. The others had a range of skills, from minimal to advanced with particular applications. Total ‘lessons,’ as at March 2009 stood at 260, given over 58 sessions. The number of individual lessons taken vary widely, from 1 to 31. From the beginning the students themselves determined the form of the lessons by quietly but very firmly indicating their desire to sit at the computer and operate it. They all made it very clear from their first session that they wanted direct control, a determination most probably linked to a desire to learn to use eMail independently. Other instruction formats had been considered and indeed planned, but were never implemented in the face of the students insistence on using the keyboard and mouse themselves, one at a time. Operationally, problems with short-term memory in this age-group would appear to dictate as much immediate hands-on experience as possible. One-to-one instruction has been continued for newcomers to the program and we are convinced that for people of this age and mental characteristics, this is the only good teaching method and the only one likely to retain their interest. Even if the environment did permit easy observation of the screen, with a video projector say, we would still advocate taking one student at a time. This has poor implications for retirement home staff helping residents to learn: one-to-one sessions are very expensive, though it is likely that volunteer help will be available (below). People in the 80+ age bracket learning to use computers require long-term support and a culture of mutual help and training amongst the residents and their family is by far the best hope for the future. All retirement complexes employ activities staff and diversional therapists, but lessons for single residents lasting around one hour at a time are not on the economic horizon. This is the key element in the need for one type of community to form if the computers are to continue to be used after university support is withdrawn. The range of interests between the Hostel residents in the first six months was very small indeed. With one exception they really only wanted access to eMail, but their interest in it was intense, though keep in mind that the group was essentially self-selected. Despite never having used eMail they all knew about it, particularly its ability to carry photographs. None initially showed any real understanding or interest in the Web or a Search Engine, although in the last six months this has substantially changed. Interest among the original group in searching for historical and genealogical information in particular is growing quite quickly, and two who joined recently are using the computers to watch back-versions of televised programs. Photographs attached to an eMail are at once the most valued part of the communication and often the most problematic. Photos that are embedded in the eMail itself, appearing automatically on the screen, only cause problems with the need to scroll a long way, something not always recognised. Opening a photo

61

ISBN: 978-972-8924-82-9 © 2009 IADIS

attached to the eMail is a problem which no Hostel resident has as yet really mastered. All have trouble identifying the attachment with its ‘paper-clip’ icon as the place to click, and the file must then be opened by another application. This second step is still causing bafflement. None of the Hostel Residents expressed any confidence that they would learn to use eMail independently, and three expressed outright doubt. From their conversation the idea that you can be too old to learn about computers does seem to be playing a significant part here, although they are all very sensitive to the problems advancing age brings to learning new things. There does seem to be a popular conviction, often expressed, that young people can and do learn to use the technology with ease. General mental ability is remarkably constant across the active period of a life-span. A long term study by Deary et el. (2000) found considerable stability between the ages of 11 and 77. There does seem to be a relation between healthy old age and decline in working (short-term) memory function (Small 2001, Small, et al. 1999) and cognitive speed (Brosseau, et al. 2007, Christensen 2001) although the extent and causes are still open to question (Foster 1999, Johnson, et al. 2002, Larrabee and Crook 1994, Rinn 1998). It certainly appears in this research that here was the outstanding difference between the retirees and younger people learning to use eMail. Things a younger person will generally find easy to retain required re-teaching week after week. The Hostel group range in age from 84 to 97, well above the age of 77 at which Deary et el. (2000) consider deterioration may be expected to set in. Indeed, all Hostel residents except one displayed the age-related mental problems above, but with no particular pattern, these problems are too variant and individual to form one in so small a group. To date, age considered, the outstanding student is the 97 year old. This is really the only serious block to the aged learning to use of the Internet independently, but note that it is not a problem with the technology as such. Differences in ability to remember new things between the Hostel and those from the Independent Units is striking: the ‘Independents,’ ages 66 to 87, retain lesson content far more easily. It is easy for an experienced user to forget the sheer volume of data displayed on a computer screen. Most of it totally ignored most of the time, but someone new to Windows must first learn what to pay attention to. Besides multiple Windows icons, the number of sensitive areas in a typical browser page is enormous. There is the information hidden in pull-down menus, and things like scroll-bars and their tiny controller icons are not only hard to see and hit with the pointer, but difficult for a beginner to control, requiring either the mouse to be repeatedly clicked, held down for a period pre-determined by experience, or dragged-and-dropped. Different sensitive areas of the screen have different ways of indicating the mouse is over them: changing colour, being underlined, flashing an I-beam, or not doing anything at all. Links are single-clicked, but icons double-clicked. The form of URLs and eMail addresses differ markedly. Passwords have their own rules. All this even a moderately experienced user knows. Setting up an eMail account consistently took an hour or even more. Two of the six original Hostel residents in the program were observed to log in to their accounts completely unaided after six lessons over 10 weeks. This does not mean they could do so consistently however. All residents taking lessons show excitement and enthusiasm. All participating Hostel residents have sent eMails and received them from family or friends. The volume of eMail varies widely, both in number and length. Undoubtedly the most satisfying part of this research is to witness the receipt of the first eMail from family, particularly eMail from grandchildren or a niece (very little, to date, from nephews). No attempt has been made to quantify this enthusiasm, but it is by far the most outstanding feature of the project. Even though the group is small, the roster for places each week usually cannot accommodate everyone who would like to be included. Essentially, other activities permitting, everyone with an eMail account would like a booking every week. The benefits to residents can also be seen in the desire and determination to access eMail independently. Sometimes they have been successful, but from observation and Staff reports, failure due to log-in problems are still significant. On one occasion at the six-month mark, two residents were reported as spending an hour and a half unsuccessfully trying to log-in to their accounts. While disappointing and frustrating to all concerned, this level of persistence does illustrate the benefit residents perceive they are deriving. For three weeks in January 2008 the computers were off-line due to problems with the telephone system. This caused an observable amount of distress within the Hostel group, emphasising the very high value they placed on the facility. As well as the temporary loss of contact itself, Staff reported they expressed concern they would forget how to use eMail if the connection was not quickly reestablished.

62


Initially, only one resident showed an interest in the World-Wide-Web. His interest was in family history and genealogy, only asking to set up an eMail account after three lessons, and then only sending a few to a friend, receiving one in return. He is the only resident so far to enquire about the structure and meaning of a URL. Three others began asking about these same Web areas after four months, and interest is now growing quite strongly among the rest of the group. Another developing area is Google Earth, both to view places the resident has lived, and where younger relatives are now living.

5. SET-UP The costs of providing an Internet connection in a retirement complex are actually coming down. eMail and Web access and operations do not require a high-performance machine, nor do there need to be many of them in any one complex. One thing retirees have is time flexibility, so even one computer goes a long way. Machines of the required standard can be had new from as little as A$500, but in a complex of any size it is probable that second hand computers will be available from relatives or volunteer groups. The main hardware expense is probably in the provision of suitable pointing devices such as trackballs (found to be much superior to mice), a printer and perhaps wireless networking equipment. Modern hardware gives little trouble and a small maintenance budget is all that is required. To make the two most popular activities, eMail and surfing the Web easier, the Firefox Homepage is set to the Hotmail log-in window, and Internet Explorer to Google. (It would have been better the other way around, because IE runs the Hotmail pages easily, but is giving trouble with some other Web applications, notably with video.) The cost of an Internet Service Provider is not great, but it is not negligible either. For the aged, the cheaper Internet connections will not do. A very reliable Broadband connection is mandatory. By far the biggest problem for seniors accessing eMail is the entry of a Username and Password (Murnane 2008) and the extra complexity of a dial-up connection with its attendant Log-in requirements and inevitable drop-outs is unacceptable, as is the down-load speed: seniors need to work within predictable time-frames. Those new to the Internet have a very strong tendency to blame themselves when things go wrong, or do not happen within the usual time interval. This leads to bafflement. Apart from recruiting eMail-experienced volunteers, the retirement facility staffs best contribution is in encouraging computer use at all times, and particularly by actively involving their families. However, if a minute of two can be spared by a passing Staff member for a resident at the computer wanting some help, it would contribute a lot. Leith Park residents comment very favorably on the help they get in this way. The overall costs involved are not negligible. Undoubtedly some retirement complexes can absorb them without trouble, but for many they represent a significant stumbling block. It is time that Government regulators factored them into their calculations.

6. TWO COMMUNITIES: RESIDENTS AND FAMILY Given this experience, it is apparent that, for the project to be successful, two very different types of communities need to flourish, one face-to-face among the Hostel residents, and one online between residents and their families and friends. Both must develop successfully or Internet use in the Hostel will probably decline when the current teaching support is withdrawn. Even when experienced Internet users begin entering supported accommodation, relatives will still need to understand the requirement to maintain close electronic links to someone whose freedom of physical movement is now very different to that which they previously enjoyed.

6.1 The Residents The form of instruction insisted on by the residents is very time-consuming and, outside a research project, very expensive, and retirement homes will generally not have the budget to employ it as part of their diversional therapy program. In the case of Leith Park, it will be the retirees themselves who will decide how

63

ISBN: 978-972-8924-82-9 © 2009 IADIS

well the experiment pans out. Essentially, a group of experienced eMailers will have to take over responsibility for teaching newcomers. After eighteen months, Staff report signs a self-sustaining culture of help is developing, but it is developing slowly. Rather naturally, residents have no need to help each other while ‘lessons’ are in progress so this behavior has not been formally observed. It is noticeable that residents not participating in lessons are showing increasing interest. The computers are set up in a wide passageway, and passers-by are increasingly stopping to chat, both about things the ‘student’ is doing, and more generally with the teacher. Staff report that outside lessons they have observed residents sitting in pairs, helping each other and sharing eMails and, particularly, photographs. Mutual help with eMail Log-in is particularly popular. When support from the University of Melbourne ceases, it is quite probable that new-comers will not get the training they need to use the Web unless someone from the existing group takes it upon themselves to provide it. In many areas, cooperation and help amongst Hostel residents is at a high level, and building a ‘community’ devoted to teaching about the Internet, within the Hostel, and perhaps with some from the Independent Living Units, is the next stage of the research program. Until now, when the subject of actually volunteering help or lessons is raised with residents it has been met with doubt. The block appears to be just lack of confidence, and as experience grows, it is hoped that an expanding circle of teaching and help will develop. It really requires a resident to volunteer to act as a ‘teaching coordinator,’ working in concert with Diversional Staff, to provide lessons and incidental help, preferably with a small group of designated helpers. Setting up an eMail account, something residents have done only once, is almost certain to require help outside the residents group. One external source of help is in volunteers. In retirement complexes in Australia, volunteers, often independently living retirees, are already routinely recruited and well organised by a designated staff member. Once equipment is installed and a program begun, attracting suitable volunteer teachers is not anticipated to be a large problem. Leith Park has two (younger) retirees who have volunteered to teach computing, but so far, over six months, this is yet to eventuate for various reasons, illustrating the multiple problems in finding people with the interest and skill who’s own changing situation allow them to participate. Latrobe University has been developing an elective in Volunteering with a view to establishing a formal subject. As part of this, three students from their career development program worked at Leith Park for eight weeks, teaching and helping with documentation and computer maintenance. The tuition was very well received. There are also groups of students from a primary and a secondary school who visit. Staff report excellent contact between the residents and the students. Coming from an educational culture which heavily emphasises computing, the students are reportedly not only eager to share their knowledge, but are apparently doing so in a manner appropriate to the circumstances. This is an area with room for considerable further study.

6.2 Family It is hard to over-emphasise the part family and friends must play in developing and sustaining interest in eMail. While there is satisfaction and stimulation in writing an eMail, this research shows it is small beside that of receiving one. No amount of formal analysis is required to authenticate the high level of excitement that accompanies receipt of a new eMail, particularly if distance has prevented a (physical) visit. Special delight at receiving a fast reply is very evident. This level of excitement and satisfaction shows no sign of diminishing after eighteen months. Photographs are particularly prized, the extra complexity involved in viewing them notwithstanding. They also form a major focus for sharing sessions with other residents. If there is excitement when a new eMail arrives, there is corresponding disappointment when no new messages appear, especially if one has been sent in the past week. In the 18 months the project has run, the number of eMails received per week by three of the residents has grown, but for the remainder it is static at best. The highest number of new replies observed at any one lesson was four. (Note that other replies may well have been opened during the week.) It could be expected that the more eMails received, the greater the enthusiasm, and this appears to be the case. The number of participating subjects is far too low to draw any real conclusions, but there does appear to be a link between the volume of eMails received and interest in logging-in.

64


It seems that families and friends require a deliberate educational program, and it is every bit as important as lessons for the residents themselves. Families must be prepared to accept responsibility and follow up messages promptly. Even a brief reply elicits a high level of excitement and satisfaction. Visitors have been invited, via the Leith Park Newsletter, to go to the computers during visits, log-in and enjoy the eMails with the resident, but to date there are only a small number reports from residents and Staff of this happening. Encouraging greater family participation calls for a very delicate approach. Gentle hints and invitations in the Newsletter may be creating awareness, but not enough response. It seems that just as we need an internal community, we also need an external one. By its nature, the individuals in a community must recognise and feel the links between them. It is not easy to find ways that a community, linked only through a relative, might characterise itself, yet some sense of belonging and contributing seems to be required. The Old Colonists’ Association strives to build a strong relationship between itself and its community of relatives and friends. Expanding this link to include a perception of the a virtual Leith Park may be one way to encourage participation. It is this characterisation that that we would like those ‘outside’ the hostel to recognise and acknowledge, because with a growing sense of community should come a growing sense of responsibility. To begin, we will suggest to families of residents without their own eMail account that they take an expanded role by creating an eMail account for the resident, and then operate it themselves while visiting. Family and friends can leave messages there knowing that the next visitor will go to the computers with the resident, read the messages and view the photographs. (Note that text while may go stale to some extent, photographs never do.) This involves family members having access to the Password, but if they are aware of the ‘group’ nature of the account that should not be a problem, and it allows any resident to join in, not just those able to use the computer autonomously. Once logged-in, residents will either be able to type their own replies or have them typed for them. Operating a ‘family’ eMail account in this way has all the advantages described above, plus the benefit of encouraging an even greater feeling of sharing within the family. It may also encourage more residents to try operating their own account. The second step will be the creation of an interactive Website, using Google Groups or similar. With the additional complications of operating a Blog, it is anticipated that most content would come from families, although some residents in the Independent Living Units are quite capable of providing material and playing Editor, and it would be an ideal place to share notes and photographs from family excursions and ‘internal’ events such as Christmas parties. Families should be able to feel part of a virtual Leith Park, even more, perhaps, than they regard themselves part of a physical one, because the virtual can be entered at will.

7. CONCLUSION In 2009 there is no room for doubt of the benefits the Internet, and particularly eMail, can bring to older retirees confined by circumstances to a supported facility. Experience at Leith Park shows that, cognitive problems and degeneration notwithstanding, a proportion of very old retirees, even those with no keyboard experience, can learn to use eMail independently, albeit over a quite extended time frame, and that rewards are significant and very worthwhile. Some sort of external help is probably required to start the process, but once an Internet culture begins to develop, two communities, one inside the retirement complex, and one external, will be needed to sustain it. Waiting until a generation of Internet users arrive with their own equipment is not good enough, and even then, it will still be up to the external community to maintain the residents’ interest for as long as absolutely possible.

ACKNOWLEDGEMENT My thanks to The Old Colonists’ Association of Victoria Leith Park residents and staff.

65

ISBN: 978-972-8924-82-9 © 2009 IADIS

REFERENCES Brosseau, J. et al., 2007. Aging affects motor skill learning when the task requires inhibitory control. Developmental Neuropsychology, Vol. 32, No. 1, pp. 597–613. Butler, G., 2005. Who's in our classrooms? Teacher, Vol. 156, pp. 6–9. Christensen, H., 2001. What cognitive changes can be expected with normal ageing? Australian and New Zealand Journal of Psychiatry, Vol. 35, No. 6, pp. 768–775. Colvin, M., 2006. Govt plans digital TV shake-up. Australian Broadcasting Commission. http://www.abc.net.au/pm/content/2006/s1591580.htm. Accessed March 16, 2006. Deary, I.J. et al., 2000. The Stability of Individual Differences in Mental Ability from Childhood to Old Age: Follow-up of the 1932 Scottish Mental Survey. Intelligence, Vol. 28, No. 1, pp. 49–55. Dobbs, D. et al., 2005. Characteristics Associated With Lower Activity Involvement in Long-Term Care Residents With Dementia. Gerontologist, Vol. 45, No. 1, pp. 81–87. Downs, M. et al., 2006. Family carers' accounts of general practice contacts for their relatives with early signs of dementia. Dementia, Vol. 5, No. 3, p. 353—374. Foster, T.C., 1999. Involvement of hippocampal synaptic plasticity in age-related memory decline. Brain Research Reviews, Vol. 30, No. 3, pp. 236–249. Hancock, G.A. et al., 2006. The needs of older people with dementia in residential care. International Journal of Geriatric Psychiatry, Vol. 21, No. 1, p. 43 ff. Hanson, E. et al., 2007. Working together with persons with early stage Dementia and their family members to design a user-friendly technology based support service. Dementia, Vol. 6, No. 3, pp. 411–434. Information Victoria. 2002. Information on State Government. State Government of Victoria. http://www.information.vic.gov.au/whatson/index.html. Accessed April 2, 2007. Johnson, M.K. et al., 2002. Second thoughts versus second looks: An age-related deficit in selectively refreshing justactive information. Psychological Science, Vol. 13, No. 1, pp. 64–67. Larrabee, G.J. and Crook, T.H., 1994. Estimated prevalence of age-associated memory impairment derived from standardized tests of memory function. International Psychogeriatrics Vol. 6, pp. 95–104. Leonard, J., 1993. Cutting to the root of ageism: The effect of mental stimulation on the elderly. September. Ohio: The Union Institute: 180 pages. Morris, M.L. and Ballard, S.M., 2003. Instructional techniques and environmental considerations in family life education programming for midlife and older adults. Family Relations, Vol. 52, No. 2, pp. 167–174. Murnane, J.S., 2007. Retirement, Mobility and the Internet. In Tatnall, A. et al. (Eds.), 2007. Education, Training and Lifelong Learning. Heidelberg, Australia: Heidelberg Press. pp 26–28. Murnane, J.S., 2008. Age, mobility and eMail. Journal of Assistive Technologies, Vol. 2, No. 4, pp. 16–25. Older people and information technology. 2005. Aging Research Online News. http://www.aro.gov.au/newsletters/ARO Newsletter June 2005.pdf. Accessed Jaunary 20, 2009. Prensky, M., 2005. Digital natives, digital immigrants. Gifted, Vol. 135, pp. 29–31. Rinn, W.E., 1998. Mental decline in normal aging: a review. Journal of Geriatric Psychiatry and Neurology, Vol. 1, No. 3, pp. 144–158. Rönnberg, L., 1998. Quality of life in nursing-home residents: an intervention study of the effect of mental stimulation through an audiovisual programme. Age and Ageing, Vol. 27, pp. 393–397. Scarmeas, N. et al., 2001. Influence of leisure activity on the incidence of Alzheimer's Disease. Neurology, Vol. 57, No. 12, pp. 2236-2242. Small, S.A., 2001. Age-related memory decline. Current concepts and future directions. Archives of Neurology, Vol. 56, pp. 360–364. Small, S.A. et al., 1999. Selective decline in memory function among healthy elderly. Neurology, Vol. 52, pp. 1392– 1399. Tran, P. and McComb, D., 2004. Internet access and healthcare information: A survey of the elderly. The Gerontologist, Vol. 44, No. 1, p. 204. Watari, K. and Gatz, M., 2002. Dementia: A cross-cultural perspective on risk factors. Generations, Vol. 26, No. 1, pp. 32–36. White, H. et al., 2002. A randomized controlled trial of the psychosocial impact of providing internet training and access to older adults. Aging & Mental Health, Vol. 6, No. 3, pp. 213–221. White, H. et al., 1999. Surfing the net in later life: a review of the literature and pilot study of computer use and quality of life. Journal of Applied Gerontology, Vol. 18, No. 3, p. 358.

66


A SOCIAL CAPITAL PERSPECTIVE OF TRAVELERS’ PARTICIPATION IN WEB-BASED SOCIAL NETWORKING COMMUNITIES Associate Professor Linda S L Lai, School of Business, Macao Polytechnic University Rua de Luís Gonzaga Gomes, Macau SAR, CHINA

ABSTRACT This paper views the development of Web-based travel communities from the social capital perspective. The objective is to offer concrete information about this new Web 2.0 socialization and communication channel which can be applied to positively affect the travel industry’s operation. Research was conducted through observations, secondary sources, participant observations, and almost a year of investigation of prominent travel-oriented social networking sites. The activities conducted by the participants in a travel community were found to be more relational and psychologically oriented. Travelers enjoy sharing stories and experiences with other members of the virtual community, hence their continuous participation in Web-based social networking communities. Moreover, the travel industry is information intensive, and with the countless options and intangible products complicating the selection and purchase process, it is often bewildering. Having trusted individuals to support the traveler’s decision is thus inevitable and is a significant consideration in strategic travel marketing and planning. KEYWORDS Social Capital, Web-based Community, Traveler, Social Network, Participant Observation, Web 2.0

1. INTRODUCTION The popularity of the Internet is transforming the technologies and practices of tourism. In the tourism industry, the Web happens to be our collective “travel square” (Wang et al., 2002) as proven by the growing number of travelers who seek online travel communities to perform their travel-related endeavors, including obtaining travel information and tips, finalizing travel transactions, forging connections with people from other places, looking for travel companions, or simply engaging in online activities for entertainment purposes at their leisure (Wang et al., 2002). Web-based travel communities merge knowledge, identity, performance, and aesthetics to impart knowledge about tourism and living in a global community; hence, these communities can be regarded as a true global knowledge network (Jensen, 2008). Web 2.0 social networking sites for travelers, including VOVO.org, Tripadvisor.com, WAYN.com, LonelyPlanet.com, VirtualTourist.com and Travelocity.com, among countless others, have been widely adopted by Internet users and offer tremendous impact potential for the tourism industry. The big question therefore is this: With all these Web-based social networking communities, how can the travel and tourism industry make full use of all of its advantages? Moreover, what are the advantages and incentives for these consumers to participate and contribute to a Web-based community? The social capital theory (Coleman, 1990; Portes, 1998) provides a framework for understanding the varied effects of online social networking on travelers’ consumer psychology. In this paper, we study the development of Web-based travel communities from the perspective of social capital. The author’s primary purpose is to offer concrete information about this new Web 2.0 socialization and communication channel which can be applied to affect the travel and tourism industry’s operation in a positive light.

67

ISBN: 978-972-8924-82-9 © 2009 IADIS

2. A SOCIAL CAPITAL PERSPECTIVE OF WEB-BASED TRAVEL COMMUNITIES The most effective approach to studying a Web-based community is to surf the Web (Yin, 2003). In this paper, research was conducted through observations, secondary sources, participant observations, and almost a year’s worth of investigation of prominent travel-oriented social networking sites such as VirtualTourist.com and WAYN.com. These social networking sites are prime examples of how Web communities experience continuous progression and adapt to the ever-changing global society.

2.1 The Theory of Social Capital A primary foundation of Web-based communities is the building and utilization of social capital. From a general perspective, social capital is defined as the resources accumulated through relationships established among individuals (Coleman, 1990). Moreover, according to Putnam (2000), it is the “collective value of all social networks and the inclinations that arise from these networks to do things for each other.” For individuals, social capital enables them to use the resources of other members belonging to the same network/s as they. These resources can take different forms, for example, useful data, the ability to organize groups, and personal relations. In addition, social capital has been connected to a range of positive social effects; some of these effects include improved public health, lower crime rates, and more effective financial markets (Adler & Kwon, 2002). What is more important, however, is that better social capital enhances the skills for mobilizing collective behaviors and allegiance to a community, among other positive outcomes (Helliwell & Putnam, 2004). In the work of Jordan and Munasib (2006), social capital is likewise considered to be motive based.

2.2 Web-based Travel Communities In 2002, the notion of a virtual tourist community was described by Wang and colleagues: “Virtual community is not an entity but rather a process defined by its members.” In essence, they removed the community from the object realm and inserted it into the action realm. Furthermore, three factors, namely, computer systems, people, and purpose, were identified as part of the virtual community (Wang et al., 2002). For the community’s users, it is its structure and the society it offers to consumers which are the key determinants to its application as a coping mechanism (Johannesson et al., 2003). Web-based social networking communities specializing in the travel and tourism industry proliferate in the Internet nowadays. These communities wield an immense influence on the said industry owing to the level of trust among their users and the consequent transfer of knowledge (i.e., word of mouth positive or negative recommendation); these subsequently influence the future decisions travelers make, even those who were not active in the community at the time the recommendation was made (Litvin, Goldsmith & Pan, 2007). According to Litvin et al. (2007), in the travel and tourism industry, a purchase cannot be properly assessed based on a tangible basis; hence, word-of-mouth recommendations are one of the most critical methods that affect a traveler’s purchasing decision.

2.3 A Framework for the Development of Online Social Capital A four-dimensional framework (see Figure 1) for the development of social capital in an online travel community has been identified using a combination of literature analysis and participant observation in online travel and tourism social networking communities.This model includes the following dimensions: (1) structural: a bridging, bonding, and linking process between participants; (2) psychological: includes enjoyment, hedonic satisfaction, fulfillment, and motivation; (3) cognitive: includes knowledge transfer and flow (or issues of tenure and historical behavior); and (4) relational: includes recognition, commitment, sharing, trust building, and friendship. The structural, cognitive, and relational dimensions have been identified from existing literature and have been refined during the process of participant observation, while the psychological dimension is based entirely on observation and construction of motivations using the psychological literature.

68


Each of the dimensions of online social capitals illustrated in Figure 1 is discussed in greater detail below. However, it should be remembered that these dimensions are highly interrelated; in fact, a single relationship or episode may be attributed to several of these dimensions or the interactions between them (Nahapiet & Ghoshal, 1998). As such, these dimensions should be considered as the facets of a single issue rather than of different issues.

Structural Capital •Bridging •Bonding •Linking

Psychological Capital •Self-efficacy •Happiness •Eudemonia •Hedonic Enjoyment

Cognitive Capital •Expertise •Flow

Travelers’ Participation in Web-based Social Networking Communities

Relational Capital •Trust •Sharing •Recognition

Figure 1. The Four Dimensions of Social Capitals Developed at Web-Based Travel Communities

2.3.1 The Structural Dimension of Online Social Capitals The structural dimension was initially described by Nahapiet and Ghoshal (1998), who based the concept on structural and relational embeddedness. Structural embeddedness is defined as the “properties of the social system and of the network of relations as a whole” (Nahapiet & Ghoshal, 1998, p. 244). In other words, structural dimension is the big-picture network of the site and the individual’s place within it. It is primarily expressed in the online travel community in bridging, bonding, and linking processes that occur between users on the site. Bridging and bonding are differentiated by Putnam (2004). He defined bridging as the creation of connections between users from different backgrounds (or who are different from one another), while bonding is the creation of connections between users having similar backgrounds. Of course, Internet communities should be very involved before there is enough knowledge between members to create more than a superficial knowledge of usual bonding or bridging factors such as gender, race, age, or profession. During the participant observation of several travel sites such as LonelyPlanet.com and TripAdvisor.com, there were few sharing episodes of this type of information, and these ordinarily occurred only when they became relevant to knowledge sharing. For example, some posts were observed which stated that women or minorities received less positive service at a given restaurant or hotel, but it was not common for posters to share this information spontaneously. Additionally, although there was a core of users in many sites, the network of other users and relationships changed rapidly due to shifts in participation by users over time. Instead, it was more common for users to develop bridging, bonding, and linking based on experience. This can in some ways be compared to the cross-cultural bonding that occurs in some multicultural immigrant communities, in which the shared experience of immigration overcomes differences that would typically trigger a bonding process (Pieterse, 2003). Linking can be seen as the relationships between individuals and groups at different social levels (Urwin, Di Pietro, Sturgis & Jack, 2008). Most social networking sites do not provide different levels of users, but the researcher did observe some informal grouping and linking based on established groups. The most obvious stratification technique, premium or paid users and free users, did not make a noticeable difference in the researcher’s observation; in fact, most users never remark on another user’s payment status even on sites

69

ISBN: 978-972-8924-82-9 © 2009 IADIS

where this could be seen. As such, the primary observed structural process was bonding, with bridging representing a smaller portion and linking being nearly absent.

2.3.2 The Psychological Dimension of Online Social Capitals The psychological dimension consists of intrinsic motivations, including self-efficacy, enjoyment, helping desires, and hedonic motivators (such as enjoyment of improved reputation). This dimension is not explicitly defined within the literature as the other three dimensions are. However, there are some signs of the dimension that has been defined as psychological within the literature. For example, one study found that the factors of self-efficacy and enjoyment affected the sharing of knowledge within an online knowledge management system (Kankanhalli, Tan & Wei, 2005). Self-efficacy is a multi-dimensional personal psychological factor that encompasses feelings of competence and efficiency surrounding a specific facet of practice or knowledge (Bandura, 1997). Self-efficacy is the central point of Bandura’s social cognitive theory, which posits that self-efficacy is the most important factor in performance. Internet self-efficacy has been shown to be a significant moderator of effective participation in online communities (Eastin & LaRose, 2000). Happiness (including the variations hedonic enjoyment and personal expressiveness) is a further component of the psychological dimension. These two variations are described as follows. Personal expressiveness, or eudemonia, can be described as the expression of the true or interior self to others, while hedonic enjoyment is the happiness brought on by getting what one wants (Waterman, 1993). It should be noted that the psychological dimension, in addition to being a separate component of the experience of online social capital building, has significant ties to other dimensions. For example, the flow aspect of the cognitive dimension is tied to happiness, enjoyment, and feelings of self-efficacy; it is essentially a specific aspect of the psychological dimension as well as tied to the cognitive experience. However, the psychological dimension can also have negative factors that impact participation such as fear of interaction, lack of Internet self-efficacy reducing the involvement of the subject, or personal conflict between users that reduce the appeal of the site through reduction of eudemonia or personal enjoyment. The researcher observed (and even in one case experienced) “flame wars” between users, in which a conflict of opinion between users grew substantially out of control as users began to more vehemently assert their point of view and to make personal attacks against other users. The researcher attributes this to a breach of the positive psychological dimension that was unexpected or too much for the individuals to deal with. This dimension is most closely connected to the cognitive and relational dimensions, and does not seem to depend strongly on the structural dimension.

2.3.3 The Cognitive Dimension of Online Social Capitals One view of the cognitive dimension is “those resources providing shared representations, interpretations, and systems of meaning” (Nahapiet & Ghoshal, 1998, p. 244). It includes expertise (transmission of knowledge, either specialist knowledge or experience based) and flow, or the cognitive experience of immersion within the community (Wu & Chang, 2005). Flow results in feelings of seamless responses, high feelings of efficiency and happiness, and involvement, and is characterized by feelings of high skill levels and challenge, focused attention, and enjoyment of interaction (Wu & Chang, 2005). The cognitive dimension includes the contribution of knowledge. However, why individuals actually contribute this knowledge is a question. Wiertz and de Ruyter (2007) established three factors that drove this participation and contribution of knowledge, including interaction propensity, community commitment, and informational value of the community itself; in other words, the individual’s structural and relational ties to the community are factors in this contribution of knowledge. This reinforces the finding of Wasko and Faraj (2005) that individuals in networks of practice contributed when they are structurally embedded, when they have knowledge or experience, and when they believe it would enhance their reputation (which in this model is built into the psychological dimension). Some information regarding this dimension can be derived from knowledge management research. In particular, researchers did not find that loss of knowledge power and image affected the sharing of knowledge in a professional knowledge-sharing environment (Kankanhalli, Tan & Wei, 2005). However, barriers have been observed in other knowledge-sharing contexts; for example, fear of criticism or insecurity regarding relevance or accuracy has been observed to inhibit knowledge sharing in online communities of practice (Ardichvili, Page & Wentling, 2003).

70


However, this lack of cognitive confidence was not always visible in the observation of these communities. There was no way to determine whether participants were holding back from participation, or if they did not see it or were not interested; thus, it was more complex to determine whether this was a significant issue. However, the vehemence of the response when inaccurate knowledge was presented does moderate toward some degree of insecurity regarding appropriate knowledge.

2.3.4 The Relational Dimension of Online Social Capitals The relational dimension can be contrasted to the structural dimension; while the structural dimension relates to mechanical linking processes, the relational dimension refers to the actual relationships that are built from these linkages (Nahapiet & Ghoshal, 1998). However, the relational dimension builds on the structural dimension, as the bonding and bridging processes and connections are required to form initial connections and to allow relational aspects such as trust and friendship to grow. A strong personal or relational connection to specific users in a given site has been acknowledged to be a factor in building social capital and in sharing information (Wasko & Faraj, 2005). Other relational issues that have been observed to affect the sharing of knowledge in a social network or knowledge management system include generalized and specific trust in other users, pro-sharing of norms and habits, and identification of the source of knowledge (Kankanhalli, Tan & Wei, 2005). The establishment of reputation, which promotes generalized trust within the site, is a considerable issue in social networking systems as a whole (Jensen, Davis & Farnham, 2002). Sites have established a variety of mechanisms to develop user reputation and trust indicators. For example, TripAdvisor uses a profile system that associates personal information as well as allows for private messaging between users (TripAdvisor, 2008). However, one feature that most travel sites are missing is an explicit reputation establishment system such as that used by SlashDot. This system allows other users to “mod up” (or down) specific comments depending on their accuracy, relevance, humor, or other characteristics; commenters that are “modded up” then receive a higher level of privileges, including increased “modding” points. This contrast demonstrates that the establishment of reputation within travelers’ social networks is considerably more informal and is arranged by the users themselves rather than imposed by the commenting system in most cases. In most cases during the observation process, the relational aspects of the site appeared to develop organically, with users first interacting superficially and then becoming more involved, including follow-up commenting, “friending” (on sites that allowed for this), and in some cases, even planning trips or meet-ups based on the site. However, individuals who spent the most time on the site did not seem to have the strongest affiliation or relational connection with other users; this is in line with observations made by Wellman et al. (2001). The relational dimension, as one that grows organically from the other three dimensions, cannot often be easily separated from expressions of this model.

3. THE USE OF ONLINE SOCIAL CAPITAL BY TRAVELERS There are four types of social capital for travelers provided by Web-based social networks: structural, psychological, cognitive, and relational. Through observational evidence collected during participation in Web-based travel communities, it was demonstrated that users in virtual communities have still formed the same kind of “friendships” that people within physical communities do, regardless of their relationship being formed within the virtual communities. In addition, cultural norms are still imposed across the online community in order to create a common social capital. Knowledge transfer and trust are the primary manifestations of online social capital, which very often lead to travel-related decisions.

3.1 Online Social Capital as Decision Support for Travelers In the case of Web-based communities, decision support is where trust and knowledge transfer intersect (Stanford-Smith & Kidd, 2000). A representation of the entire decision support process can be gleaned from the researcher’s personal observations of the synergy and trust established among these sites, particularly a question regarding tour operators in Singapore on one of the observed site’s bulletin boards. The opening sentence of the post stated, “This looks good on the site but I wanted to see what you guys really thought,” thus revealing the lack of trust in e-commerce entities by the questioner. The person who posted the query

71

ISBN: 978-972-8924-82-9 © 2009 IADIS

then interacted with each of the respondents, exhibiting different levels of perceived trust in their replies depending on both tangible (i.e., Were they offering first-hand knowledge?) and intangible (i.e., How often they had seen the respondent’s post or comment before, or their perceived level of community involvement) factors. The questioner and the respondents both established bridging and bonding social capital as they interacted and exchanged knowledge about the process. Finally, the knowledge transfer that occurred between the respondents provided support for the final decision of the initiator of the “conversation,” which was to not patronize the tour operator because it garnered negative comments from several of the trusted posters, regardless of the existence of a few dissenting opinions. Hence, trust and other factors of online social capital had a significant role in the novice traveler’s decision-making process.

3.2 The Social Dimension of Travel and Tourist Decisions Social capital is especially beneficial to Web-based travel communities because tourism transpires in the social dimension. According to MacCannell (1976), tourism is considered one of the most significant modes of social action and is exemplary of the conditions of the post-modern man. Social dynamics as an important part of being a tourist has already been confirmed (Urry, 1990). Before going on a trip, we seek often advice from family and fiends. Likewise, when on the road, we frequently forge relationships and make acquaintance with fellow travelers. Finally, upon returning home, we want to share our experiences, positive and negative, with others; therefore, we remember the experiences and establish our identity as a traveler and as a human being. Online travel communities appear within this context. Nowadays, they are consistently used before departure to gather information and perhaps have a traveling companion, during travel to search for further information and make contact with those at home, and after travel to share photographs and tales from the travel. Individuals’ socialization process is an essential but often overlooked aspect in the study of travel and tourism decisions (Sirakaya and Woodside, 2004). In particular, the final destination may not be an initial primary concern for the traveler as long as he/she is with friends; thus, the so-called friend/s (or virtual friend) is/are permitted to make the final decision regarding the travel destination. This social pursuit usually involves family, friends, relatives, and even online friends who are linked with the traveler through the online social networking community. As previously stated, Web 2.0 social networking sites play the role of a “travel square” (Wang et al., 2002) for travelers, enabling them to network with similarly minded individuals to support their decision-making process.

4. SOCIAL CAPITAL SPECIALLY USEFUL TO WEB-BASED TRAVEL COMMUNITIES The model depicted in Figure 1 can be used to describe any online social network, as participants within most such networks will experience similar relationship formations, network placements, and other issues (modulo small changes such as formal versus informal reputation establishment, etc.) However, there are specific differences between travelers’ social networks and other social networks that should be explored in order to understand how the unique structure, focus, and practice of travelers’ social networks have evolved. The specific differences between these models include the weak structural ties paired with strong relational factors including trust, sharing, and enjoyment, as well as the importance of intrinsic returns.

4.1 Weak Structural Ties and Strong Relational Factors As noted above, there are weak structural ties between travelers’ online social networks, which the researcher believes is due to the absence of cues ordinarily used for bridging and bonding and most sites not having explicit user levels or clusters, thus reducing the impact of linking. However, even though relational capital often relies on structural capital in order to develop, relational capital within travelers’ network is very important, possibly far more important than that within other social networks. Trust between users is one of the elements of a social networking site which distinguishes it from a corporate review site that pays for its reviews (Wu & Chang, 2005). There are a number of potential reasons for this that the researcher observed, but one of the most common ones discussed for this level of trust was the inability of individuals to verify

72


information (due to travel costs, time restrictions, and so on), so accurate and trustworthy information was placed at a premium. As such, the development of personal trust throughout these networks is essential to ensure that users have access to appropriate information.

4.2 The Importance of Intrinsic Returns Another distinction between travelers’ social networks and other communities of practice or social networks is the happiness (eudemonia and hedonic enjoyment) that the users expect of the network on which they placed a high priority. Simply put, people enjoy talking about travel–remembering their past adventures or planning new ones excites them and offers an opportunity for self-expression and fulfillment. Even users who are not immediately planning a trip seem to enjoy interacting with individuals on the site and discussing issues regarding travel. The researcher believes that participants in these social networks for the most part regarded travel as a hobby rather than as a mere experience sharing; although some users were observed to offer reviews of individual services, they did not engage with the network despite offering information and did not appear to experience this effect. This corresponds more to previous research on professional knowledge sharing than to the enjoyment-based travel subject matter (Wu & Chang, 2005). A related issue is the intrinsic motivation for participation within this network as compared to others. In many cases, experienced travelers were observed to help novices eagerly, not out of any extrinsic reward system (this enthusiasm went unremarked in most cases and of course had no monetary or other extrinsic reward), but out of an apparent desire to help and a simple enjoyment of the subject matter. This is in opposition to previous research that was conducted in the professional community of practice context in which extrinsic rewards such as acknowledgment or financial or advancement goals, or even job requirements were factors in the participation in the social network (Kankanhalli, Tan & Wei, 2005; Wu & Chang, 2005). This is also related to the self-efficacy involvement embedded in the psychological dimension. The researcher believes that the difference between intrinsic and extrinsic participation motivations may be one of the most essential differences in travelers’ social networks compared with other social networks.

5. CONCLUSION AND RECOMMENDATIONS In this study, a social capital perspective was adopted to conceptualize the involvement of travelers in Webbased social networking communities, and a four-dimensional framework was forwarded to rationalize the development of social capital in these virtual communities. During the study, it was observed that the activities conducted by participants in a travel community were more relational and psychologically oriented. Travelers enjoy sharing their stories and travel experiences with other members of the virtual community, hence their continuous participation in the Web-based social networking communities. The implications of social networking and the utilization of social capital for decision support are of utmost importance to the travel and tourism industry. This industry is information intensive, and oftentimes, navigating its ins and outs turns out to be a bewildering experience, what with the countless options and intangible products complicating the selection and purchase process. Having trusted individuals to support the traveler’s decision-making process is thus inevitable, and it is a significant consideration in strategic travel marketing and planning. The rise of Web-based social networking sites such as TripAdvisor.com and LonelyPlanet.com has magnified this influence a hundredfold. Travel information is no longer passed from person to person; instead, hundreds (even thousands) of free-riders can likewise access this information by viewing the conversation. Conversation archiving and search engines can increase this influence even farther. Thus, the travel and tourism industry must avoid negative experiences as much as possible, as well as actively promote and support the spread of positive experiences using these information-sharing mechanisms. Furthermore, there should be extensive acknowledgement of trust’s role in e-commerce. Although a commercial offering of structural information such as costs and fees, booking times, and other information may be accepted by the members, corporate reviews will be regarded with a certain level of misgiving. Sites that have structural travel offerings (i.e., hotel bookings) may be better off incorporating social networking principles into their sites by providing value-added features (i.e., reviews and guest rankings). Such information will be viewed with less suspicion by users than commercial information, and will improve the site’s trust profile by allowing it to maximize the establishment of a social capital.

73

ISBN: 978-972-8924-82-9 © 2009 IADIS

REFERENCES Adler, P. and Kwon, S. (2002). Social Capital: Prospects for a New Concept. Academy of Management Review, 27 (1), 17-40. Ardichvili, A., Page, V., & Wentling, T. (2003). Motivation and barriers to participation in virtual knowledge-sharing communities of practice. Journal of Knowledge Management , 7 (1), 64-77. Bandura, A. (1997). Self-efficacy: The exercise of control (7th ed.). New York: MacMillan. Coleman, J. (1990). Foundations of Social Theory. Boston, MA: Harvard University Press. Eastin, M., & LaRose, R. (2000). Internet self-efficacy and the psychology of the Digital Divide. Journal of ComputerMediated Communication , 6 (1). Helliwell, J. F. & Putnam, R. D. (2004). The Social Context of Well-Being, Philosophical Transactions of the Royal Society, 359 (1449), 1435-1446. Jensen, C., Davis, J., & Farnham, S. (2002). Finding others online: Reputation systems for social online spaces. Proceeedings of CHI 2002 (pp. 447-454). Minneapolis: ACM. Jensen, J. (2008). Virtual Tourist: Knowledge communication in an online travel community. International Journal of Web Based Communities, 4 (4), 503-522. Johannesson, G., Skaptadottir, U. & Benediktsson, K. (2003). Coping with social capital? The cultural economy of tourism in the North. Sociologica Ruralis 43(1), 3-16. Jordan, J. L. & Munasib, A. B. A. (2006). Motives and social capital consequence. Journal of Economic Issues XL (4), 1093-1112. Kankanhalli, A., Tan, B., & Wei, K. (2005). Contributing knowledge to electronic knowledge repositories: An empirical investigation. MIS Quarterly , 29 (1), 113-143. Litvin, S.W., Goldsmith, R.E. & Pan, B. (2007). Electronic word-of-mouth in hospitality and tourism management. Tourism Management 28, 1-11. MacCannell, D. (1976). The Tourist. A New Theory of Leisure Class. New York: MacMillan. Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage. The Academy of Management Review , 23 (2), 242-266. Pieterse, J. (2003). Social capital and migration: Beyond ethnic economies. Ethnicities , 3 (1), 29-58. Publications, Inc. Putnam, R. D. (2004). Democracies in flux: The evolution of social capital in contemporary society. New York: Oxford University Press US. Putnam, R.D. (2000). Bowling Alone. New York: Simon & Schuster. Sirakaya, E. & Woodside, A.G. (2005). Building and testing theories of decision making by travellers. Tourism Management 26, 815-832. Stanford-Smith, B. & Kidd, P.T. (2000). E-Business: Key issues, applications and technologies. Oxford: IOS Press. TripAdvisor. (2008). What's my profile? Retrieved January 28, 2009, from Help Center: http://www.tripadvisor.com/help/whats_my_profile Urry, J. (1990). The Tourist Gaze. London: Sage. Urwin, P., Di Pietro, G., Sturgis, P., & Jack, G. (2008). Measuring the returns to networking and the accumulation of social capital: Any evidence of bonding, bridging, or linking? American Journal of Economics and Sociology , 67 (5), 941-968. Wang, Y., Yu, Q. & Fesenmaier, D. (2002). Defining the virtual tourist community: implications for tourism marketing”, Tourism Management 23, 407-417. Wasko, M., & Faraj, S. (2005). Why should I share? Examining social capital and knowledge contribution in electronic networks of practice. MIS Quarterly , 29 (1), 35-57. Waterman, A. (1993). Two conceptions of happiness: Contrasts of personal expresiveness (Eudaimonia) and hedonic enjoyment. Journal of Personality and Social psychology , 64 (4), 678-691. Wellman, B., Haase, A., Witte, J. & Hampton, K. (2001). Does the Internet increase, decrease, or supplement social capital? Social networks, participation and community commitment. American Behavioral Scientist , 45 (3), 435-455. Wiertz, C. & de Ruyter, K. (2007). Beyond the call of duty: Why customers contribute to firm-hosted commercial online communities. Organization Studies , 28, 347-376. Wu, J. & Chang, Y. (2005). Towards understanding members' interactivity, trust, and flow in online travel community. Industrial Management + Data Systems , 105 (7), 937-954. Yin, R. K. (2003). Case Study Research: Design and Methods. California: Sage

74


KNOWLEDGE MANAGEMENT AND SHARING IN OPEN SOURCE COMMUNITIES S. L. Toral and F. Barrero E. S. Ingenieros, University of Seville * Avda. Camino de los Descubrimientos s/n, 41092, Seville, Spain

M. R. Martínez-Torres E. U. Estudios Empresariales, University of Seville Avda. San Francisco Javier, s/n 41018 Sevilla (Spain)

ABSTRACT Open Source Communities are typically supported by hundreds o even thousands of developers and users geographically spread all over the world. Participants get in contact using electronic media like CVS, mailing lists and forums in which they can post source code, bugs, problems, doubts or experiences and receive answers, solutions or alternative from other community users. This scheme leads to a structure in which learning takes place as a social process associated to practice and knowledge is captured in the form of code or discussions. This aim of this paper consists of analyzing the stored information to observe the knowledge creation process and its evolution using a case study. Latent semantic analysis techniques will be applied for this purpose. The obtained results will provide new insights about knowledge management in virtual communities. KEYWORDS Virtual communities, open source, knowledge creation, latent semantic analysis, social learning.

1. INTRODUCTION The free software and open source software phenomenon is one of the most significant discontinuities in the field of Information and Communication Technologies in the sense of raising new ways of doing in areas like R&D, marketing and distribution or income generation models (Korica et al., 2006). One of the most exciting topics around Open Source Software (OSS) projects is how collaboration is created and maintained over time. Hundreds or even thousands of developers geographically distributed get in contact and collaborate to develop and improve certain underlying software, using coordination mechanisms far away from the ones used in traditional software projects. In many cases, the results are even more successful than proprietary counterparts, as in the case of web servers (Apache) or scripting languages (PHP), or at least represent a serious competence to proprietary software, like in the case of operating systems (GNU/Linux), web browsers (Mozilla) or database management systems (MySQL). One of the bases of this community development model is social learning theory (Wenger, 1998). It postulates that learning is a social process (Henri & Pudelko, 2003), situating learning in the context of the social experience of individuals. The major premise for this theory is that human beings are social beings and the context in which learning takes place are defined as communities of practice. From the perspective of the social learning theory, practice can be defined as “. . . the way tasks are done, spontaneous, improvised, responding to a changing, unpredictable environment, driven by tacit knowledge” (Seely Brown and Duguid, 2000). Learning requires the participation of individuals inside a community, as a difference to the traditional idea of considering learning as an individual process and as a result of a teaching. This is the case of open source communities, where people can freely post their questions related to the underlying software and receive some solutions or alternatives from someone else of the community. Knowledge is then acquired as a result of the interactions among people through the discussions about a particular topic. These discussions are

75

ISBN: 978-972-8924-82-9 © 2009 IADIS

frequently stored and they are publicly available, son people can consult previously discussed topics and find an answer to a particular problem. The purpose of this paper consists of analyzing this stored information using latent semantic analysis techniques to understand the processes of knowledge management involved in the community members’ interactions. The rest of the paper is organized as follows: Section 2 is devoted to knowledge management in virtual communities. In particular, the notion of communities of practice and its relation with knowledge management as well as the methodology for the data treatment will be described. Section 3 details the case study focused on a Debian Linux port to an embedded processor. Finally, conclusions are summarized.

2. KNOWLEDGE MANAGEMENT IN VIRTUAL COMMUNITIES Virtual communities are frequently analyzed in terms of communities of practice (Ardichvili et al., 2003; Zhang and Watts, 2008), if they satisfy the three dimensions defined by Wenger (1998). In this section, we will give a brief overview about communities of practice and how they can be used as an approach to knowledge management. Finally, the methodology to characterize one of the processes related to knowledge management will be detailed.

2.1 Communities of Practice Virtual communities can be studied from the perspective of Communities of Practice (CoP) developed by Lave & Wenger (1991). This concept refers to the process of social learning that occurs when people who have a common interest in some subject or problem collaborate over an extended period to share ideas, find solutions, and build innovations. CoPs are not formal structures, such as departments or project teams. Instead, they are informal entities, which exist in the mind of their members, and are glued together by the connections the members have with each other, and by their specific shared problems or areas of interest (Wenger and Snyder, 2000; Ardichvili et al., 2003). According to Wenger (1998), a CoP is defined through three dimensions: • Mutual engagement, which describes how community members interact with each other in the practice. • Joint enterprise, which embodies the shared interest of community members and the goal of the community as a whole, and symbolizes what the community is about. • Shared repertoire, which consists of ‘‘routines, words, tools, ways of doing things, stories, gestures, symbols, genres, actions, or concepts that the community has produced or adopted in the course of its existence, and which have become part of its practice’’ (Wenger, 1998). Virtual communities use networked technology, especially the Internet, to establish collaboration across geographical barriers and time zones (Johnson, 2001). In contrast to traditional communities, virtual communities require less formal borders and norms do not dominate as much as in traditional communities, because members cannot see each other. The question about the similarities between face-to-face communities and virtual communities is currently open in the information science literature (Hersberger et al., 2007). Johnson (2001) argues that a virtual community is the designed community, whereas the cop is what emerges from the designed community. Carter (2005) indicates that virtual relationships are no more exotic than traditional relationships, as virtual relationships are quickly assimilated into everyday life. Zhang and Watts (2008) emphasize it is much easier to join online communities than conventional CoPs, with the additional possibilities of one-to-many or even many-to-many communications. Graham and Hazel (2004) highlight the role of membership in virtual communities as a critical factor in determining the level of knowledge capital generated by an online group. Although the majority if authors agree that CoPs can exist online, there are limitations to virtual communities. For instance, free riders (Dyer and Nobeoka, 2000) represent a serious problem respect to knowledge sharing activities. Free riders can be defined as “members who enjoy the benefits of the collective good without contributing to its establishment”, and they are also known as lurkers (Millen, Feinberg, and Kerr, 2005; MclureWasko and Faraj, 2005. Another limitation is motivating network members to participate in the community and to share valuable knowledge with each other openly (Dyer and Nobeoka, 2000). Bradshaw, Powell, and Terrel (2004) state that

76


the development of distributed CoPs may take time and requires constant checking of objectives and working practices to keep the community working properly.

2.2 Knowledge Management Several prior studies suggest the suitability of CoPs as a Knowledge Management (KM) tool (Liedtka, 1999; Philips and Bonner, 2000). CoPs represent an approach to knowledge management focused on knowledge and knowing in practice (Lave and Wenger, 1991; Wenger, 1998). They provide an environment for people to develop knowledge through interaction with others in an environment where knowledge is created, nurtured, and sustained (Hildreth y Kimble, 2002). Allee (2000) points out that the community of practice is an intrinsic condition for knowledge to exist, since it cannot be separated from the group that creates it, uses it and transforms it. Fowler and Mayes (1999) highlights that value is given by social participation, in particular, by being an active participant in the practices of social communities, and by constructing an identity in relation to each community. Most of the current approaches to KM are based on the division of knowledge into dichotomous opposites such as the tacit-explicit distinction popularized by Nonaka and Takeuchi (1995). Tacit knowledge is the implicit knowledge used by organizational members to perform their work and to make sense of their worlds (Choo, 1998). Tacit knowledge is interwoven with the context and the experiences of its respective ‘owner’ and therefore, hard to transfer (Russel, Sambamurthy, and Zmud, 2001). On the contrary, explicit knowledge can be codified and is therefore easy to transfer. This tacit/explicit knowledge dichotomy has received wide acceptance in organizational and scientific literature, because it recognizes the intangible, human related side of knowledge. However, some authors have argued that knowledge should not be regarded as a dichotomy, but as a duality (Hildreth and Kimble, 2002), They claim that that all knowledge has both harder and softer aspects. The harder aspects of knowledge as those that can be made explicit, are structured, codified and can be captured and stored in knowledge repositories. The softer aspects of knowledge are those aspects that are less structured and are difficult or impossible to articulate (Kimble and Hildreth, 2005). It is the implicit knowledge that is embedded in peoples’ everyday experiences and actions (Wenger, 1998; Hildreth et al., 1999; Seely Brown and Duguid, 2000). As a difference to Nonaka and Takeuchi (1995), Kimble and Hildreth (2005) consider hard and soft knowledge as a duality rather than a dichotomy. This vision emphasizes that both sides of knowledge must be taken into account if an attempt to manage knowledge. The process underlying the construction and nurturing of soft knowledge in CoPs is called Legitimate Peripheral Participation (LPP) (Lave and Wenger, 1991). LPP describes the process by which a newcomer is integrated into the community. In this process, new members learn how to function as a community member through participation, and acquire the language, values, and norms of the community. Learning is gradually achieved as an individual moves from being a novice, gaining access to community practices to complete socialization and therefore becoming an insider or full member of the community. For instance, OSS projects web sites provide forums and mailing lists where participants and contributors can report software improvements, needs or bugs, and share and discuss solutions to posted messages. The other half of the duality is reification, which means giving concrete form to something that is abstract. It is the process underlying the construction of hard knowledge. Both processes are developed together in CoPs and they affect the way in which meaning is negotiated. Participation and LPP process can be analyzed using social network analysis techniques. But this paper will be focused on the dual process, that is, reification. In the case of OSS projects, the harder aspect of knowledge is stored and publicly available through forums, discussion and repositories. The analysis of the reification process is the analysis of this information. In this paper, latent semantic analysis techniques will be used to extract the main topics treated each year, and their evolution over time. The proposed methodology will be detailed in the next subsection.

2.3 Methodology Modern semantic analysis techniques are based on a vector space model (Salto and McGill, 1983), in which documents are summarized and represented by vectors of words (term vectors). However, the main problem of such representation is the high dimensionality of the feature space (one dimension for each unique word). Therefore, it is desirable to first project the documents into a lower-dimensional subspace in which the

77

ISBN: 978-972-8924-82-9 © 2009 IADIS

semantic structure of the document space becomes clear (Cai et al., 2005). In the low-dimensional semantic space, the traditional clustering algorithms can be then applied. To this end, spectral clustering (Shi and Malik, 2000; Ng et al., 2001), clustering using Latent Semantic Indexing (Zha et al., 2001), and clustering based on nonnegative matrix factorization (Xu et al., 2003; Xu and Gong, 2004) are the most well-known techniques. Different to LSI, in this paper we have applied the topic model, which is a statistical language model that relates words and documents through topics. It is based upon the idea that documents are mixtures of topics, where a topic is a probability distribution over words (Blei et al., 2003; Hofmann, 2001). Representing the content of words and documents with probabilistic topics has one distinct advantage over the purely spatial representation of LSI. Each topic is individually interpretable, providing a probability distribution over a word that picks out a coherent cluster of correlated terms. Hofmann (2001) introduced the probabilistic topic approach to document modelling in his Probabilistic Latent Semantic Indexing method (pLSI). Blei et al. (2003) extended this model by introducing a Dirichlet prior, calling the resulting generative model Latent Dirichlet Allocation (LDA). In both cases, the basic idea is that documents are represented as random mixtures over latent topics, where each topic is characterized by a distribution over words.

3. CASE STUDY The proposed case study is based on Linux, which is one the most famous and cited OSS projects. Among the Linux distributions, Debian is one of the most well-known. The Debian Project, which was founded in 1993 by Ian Murdock, is a worldwide group of volunteers who endeavor to produce an operating system distribution that is composed entirely of free software (Michlmayr and Senyard, 2006). Debian GNU/Linux software distribution, which includes the Linux operating system kernel and thousands of prepackaged applications, is developed through distributed development all around the world. Much of the conversation between Debian developers and users is managed through several mailing lists, which can be easily accessed through the Debian web site (http://lists.debian.org/ ). Among the numerous mailing list available, we have chosen a Linux port to non-i386 Linux architecture: debian-arm, which is the Debian port to ARM architecture. There are several reasons justifying this choice: • Linux is clearly the most used operating system for non-i386 processor architectures (Yaghmour, 2003). • In contrast to other typical open source projects or even desktop Linux projects, most contributions in Linux ports to non-conventional architectures do not come from volunteers or hobbyists, but from commercial firms, many of which are dedicated embedded Linux firms. This fact facilitates the analysis of the discussions and the knowledge management processes. • Among non-conventional architectures, ARM is the industry's leading provider of 32-bit embedded RISC microprocessors, offering a wide range of processors based on a common architecture and delivering high performance together with low power consumption and system cost. Today, a lot of smart phones, GPS navigations devices, PDAs, media players, digital cameras and TVs or video consoles are using an ARM inside (Lamie, 2005). The Debian port to ARM mailing list (http://lists.debian.org/debian-arm/) has been analyzed during its lifetime from 1999 to 2007. Up to 7482 different messages were considered. They have been downloaded and processed using specific software developed using MATLAB®. MATLAB® is a high-level language and interactive environment that enables you to perform computationally intensive tasks faster than with traditional programming languages such as C, C++, and Fortran. The sequence of steps followed to process the information is next detailed. All of them are repeated for each of the considered years. 1. The developed program downloads all the messages corresponding to each year, and then extracts the header and the body of the message. In particular, HTML tags are processed for this purpose. 2. Punctuation marks, HTML tags and prepositions are removed. Afterward, all the words are lower case converted. Finally, the frequency-of-occurrence-rates for word tokens are obtained. 3. The most repeated words are used as the basic vocabulary to describe the domain of study. 4. The final step is the topic extraction using Latent Dirichlet Allocation algorithm. The Latent Dirichlet Allocation algorithm involves the selection of several parameters to achieve a topic description of the domain under study.

78


Table 1. Dimensions involved in LDA algorithm. Parameter D W N L T ITER

Description Number of documents in corpus Number of words in vocabulary Total number of words in corpus Average length of document in words (L = N/D) Number of topics Number of iterations

Error! Reference source not found. details the key dimensions or size parameters describing a corpus (D, W, N and L) and a topic model run (T and ITER). In our case study, each year is considered a different corpus. The aim is extracting topics for each year and compares the topic evolution over time. Error! Reference source not found. details the particular parameters of the debian ARM mailing list during the considered period of time Table 2. Dimensions for the Debian Linux ARM mailing list. Year 1999 2000 2001 2002 2003 2004 2005 2006 2007

D 552 667 956 848 377 684 625 1098 1675

W 410 410 410 410 410 410 410 410 410

N 100339 115869 193313 218035 82904 113440 108523 218565 356481

L 181,77 173,72 202,21 257,12 219,90 165,85 173,64 199,06 212,82

Although the number of documents (D column) exhibits high variations among years, the average length of documents in words (L column) show slighter variations. The most active years are the two last ones, 2006 and 2007, which is concordant with the growing interest in Linux ports to embedded processors. Two parameters must be selected before running the algorithm, as the number of topics or the number of iterations. The number of Gibbs sampler iterations has been chosen ITER = 200. This is a large enough value to guarantee the convergence of the algorithm (Blei et al., 2003). The number of topics has been selected using perplexity. Perplexity is a standard measure of performance for statistical models of natural language (Manning & Schutze, 1999), and it is defined by Eq. (1). W ⎞ ⎛ 1 pplex = exp⎜ − log P ( wn | d n ) ⎟ (1) n = 1 ⎠ ⎝ W The results of running the topic model during the years 1999-2007 is detailed in Error! Reference source not found.. In particular, the resulting topics for each year are detailed including the five more relevant words for each topic. The obtained results can used to characterize the reification process underlying the construction of the hard knowledge. Three specific types of knowledge sharing categories can be distinguished. The first one consists of revealing personal uniquely acquired experience and knowledge. Once the knowledge is revealed, it can be reused in other fields or re-applied in different practices. Knowledge reusing avoids duplication of effort in re-inventing solutions (Wai, 2008). Finally, knowledge can be recombined to generate new knowledge (Kuk, 2006). These three categories are embedded in the reification process and they can be measured using indirect indicators: • The number of topics (obtained as the number of topics minimizing the perplexity of equation 1) varies form one year to another. In the analyzed period of time, the minimum value is 9 and corresponds to years 2000 and 2004, while the maximum value is 19, corresponding to year 2003. The number of topics is a measure of the variety of discussions and the knowledge revealed. Notice that no topic is exactly repeated during the analyzed period of time. That means that each topic means a new field in which practice is revealed. • Although there are no topic repetition, topics show words in common with some other topics. This is one of the advantages of the proposed methodology, its capability of dealing with words that can be used in

∑

79

ISBN: 978-972-8924-82-9 © 2009 IADIS

•

different contexts. For instance, the word kernel appears 10 times, but in different contexts related to compilation, configuration or availability issues. Polysemy is related to knowledge reuse processes. Finally, messages can be assigned to one or more of the identified topics. Those messages owing to several topics can be used as a measure of the knowledge recombination. Table 3. Topic distribution per year

80


4. CONCLUSION This paper deals with knowledge sharing in open source communities. In particular, the study is focused on the reification process associated to hard knowledge development. For this purpose, a latent dirichlet allocation algorithm has been used to extract the main topics of the Debian ARM Linux mailing list during the period of time 1999-2007. The obtained results allow defining a set of indicators measuring the three categories of knowledge sharing, that is, knowledge revealing, reuse and recombination. The main limitation of this paper is using just one case study. As a future, work, more communities can be studied and analyzed, by applying a similar methodology.

81

ISBN: 978-972-8924-82-9 © 2009 IADIS

ACKNOWLEDGEMENT This work has been supported by the Spanish Ministry of Education and Science (Research Project with reference DPI2007-60128) and the Consejería de Innovación, Ciencia y Empresa (Research Project with reference P07-TIC-02621)

REFERENCES Allee, V. 2000, Knowledge networks and communities of practice, OD Practitioner, Vol. 32 No. 4, pp. 1-15. Ardichvili, A., Page, V., Wentling, T. 2003. Motivation and barriers to participation in virtual knowledge-sharing communities of practice, Journal of Knowledge Management; Vol. 7, Iss. 1, pp. 64-77. Blei, D. M., Ng, A. Y., and Jordan, M. I., 2003. Latent Dirichlet Allocation, Journal of Machine Learning Research, Vol. 3, pp. 993-1022. Bradshaw, P., Powell, S., and Terrel, I. 2004. Building a community of practice: Technological and social implications for a distributed team. In Hildreth, Paul M. and Kimble, Chris, editors, Knowledge Networks: Innovation through Communities of Practice, pp. 184–201. Idea Group Publishing. Cai, D., He, X., and Han, J., 2005. Document Clustering Using Locality Preserving Indexing, IEEE Transactions on Knowledge and Data Engineering, Vol. 17, no. 12, pp. 1624-1637. Carter, D., 2005, “Living in virtual communities: an ethnography of human relationships in cyberspace”, Information, Communication & Society, Vol. 8 No. 2, pp. 148-67. Choo, C. W. 1998. The knowing organization: How organizations use information to construct meaning, create knowledge and make decisions. Oxford University Press, New York, USA. Dyer, J. H. and Nobeoka, K. 2000. Creating and managing a high-performance knowledge-sharing network: the toyota case. Strategic Management Journal, 21, pp. 345–367. Graham, D. and Hazel, H. (2004), “Creation and recreation: motivating collaboration to generate knowledge capital in online communities”, International Journal of Information Management, Vol. 24 No. 3, pp. 235-46. Henri, F., & Pudelko, B., 2003. Understanding and analysing activity and learning in virtual communities. Journal of Computer Assisted Learning, Vol. 19, Iss. 4, pp. 474 – 487. Hersberger, J. A., Murray, A. L., Rioux, K. S., 2007. Examining information exchange and virtual communities: an emergent framework, Online Information Review, Vol.:31, Iss. 2, pp. 135 – 147. Hildreth, P., Wright, P., and Kimble, C. 1999. Knowledge management: Are we missing something? In Brooks, L. and C., Kimble, editors, Information Systems - The Next Generation, York, pp. 347–356. Hildreth, P.M. and Kimble, C. 2002. The duality of knowledge. Information Research, Vol. 8, Iss. 1, pp. 1–17. Hofmann, T., 2001. Unsupervised Learning by Probabilistic Latent Semantic Analysis, Machine Learning Journal, Vol. 42, no. 1, pp. 177-196. Johnson, C. M. 2001. A survey of current research on online communities of practice. Internet and Higher Education, 4, pp. 45–60. Kimble, C., and Hildreth, P., 2005. Dualities, distributed communities of practice and knowledge management. Journal of Knowledge Management, Vol. 9, no. 4, pp. 102-113, Korica, P., Maurer, H., Schinagl, W., 2006, The Growing Importance of e-Communities on the Web. Proc. of the Intl. Conference on Web Based Communities, IADIS 2006, San Sebastian, España, pp. 165-174. Kuk, G. 2006. Strategic Interaction and Knowledge Sharing in the KDE Developer Mailing List. Management Science, Vol. 52, No. 7, pp. 1031–1042. Lamie, E. L., 2005. Real-Time Embedded Multithreading using ThreadX and ARM. CMPBooks, CA, Lave J. and Wenger E., 1991. Situated learning: Legitimate peripheral participation. Cambridge University Press. Liedtka, J., 1999. Linking Competitive Advantage with Communities of Practice, Journal of management Inquiry, Vol. 8, no. 1, pp. 5-16. Manning, C. D. & Schutze, H., 1999. Foundations of Statistical Natural Language Processing, MIT Press, Cambridge, MA. Mclure Wasko, M. and Faraj, S. 2005. Why should i share? examining social capital and knowledge contribution in electronic networks of practice. Management Information Systems Quarterly, 29, 1, pp. 35–57. Michlmayr, M., and Senyard, A. 2006. A Statistical Analysis of Defects in Debian and Strategies for Improving Quality in Free Software Projects. In The Economics of Open Source Software Development, Jürgen Bitzer and Philipp J. H. Schröder (eds.), pp. 131-148.

82


Millen, D., Feinberg, J., and Kerr, B. 2005. Social bookmarking in the enterprise. ACM Queue, Vol. 3, No. 9, pp. 28–35. Ng, A.Y., Jordan, M., and Weiss, Y., 2001. On Spectral Clustering: Analysis and an Algorithm, Advances in Neural Information Processing Systems 14, pp. 849-856, Cambridge, Mass.: MIT Press. Nonaka, I. and Takeuchi, H. 1995. The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation. Oxford University Press, New York, USA. Philips, J., and Bonner, P.D. 2000. Motivation, Knowledge Transfer, and Organizational Forms, Organization Science, Vol. 11, No. 5, pp. 538-550. Russel, L. P., Sambamurthy, V., and Zmud, R. W. 2001. The assimilation of knowledge platforms in organizations: An empirical investigation. Organization Science, Vol. 12, no. 2, pp. 117–135. Salto, G., and McGill, M.J., 1983. An Introduction to Modern Information Retrieval, New York: McGraw-Hill. Seely Brown, J. and Duguid, P. 2000. Balancing act: How to capture knowledge without killing it. Harvard Business Review, Vol. 78, Iss. 3, pp. 73–80. Shi, J., and Malik, J., 2000. Normalized Cuts and Image Segmentation, IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 22, no. 8, pp. 888-905. Wai, F.B. 2008. Reuse of knowledge assets from repositories: A mixed methods study. Information & Management, Vol. 45, Iss. 6, pp. 365-375 Wenger, E., 1998. Communities of Practice: Learning, Meaning and Identity. Cambridge University Press, New York, USA. Wenger, E.C. and Snyder, W.M., 2000, ‘‘Communities of practice: the organizational frontier’’, Harvard Business Review, Vol. 78 No. 1, pp. 139-44. Xu, W., Liu, X., and Gong, Y., 2003. Document Clustering Based on Non-Negative Matrix Factorization, Proc. Int’l Conf. Research and Development in Information Retrieval, pp. 267-273. Xu, W., and Gong, Y., 2004. Document Clustering by Concept Factorization, Proc. Int’l Conf. Research and Development in Information Retrieval, pp. 202-209. Yaghmour, K., 2003. Building Embedded Linux Systems. O'Reilly. Zha, H., Ding, C., Gu, M., He, X., and Simon, H. 2001. Spectral Relaxation for k-Means Clustering, Advances in Neural Information Processing Systems 14, pp. 1057-1064, Cambridge, Mass.: MIT Press. Zhang, W., Watts, S., 2008. Online communities as communities of practice: a case study. Journal of Knowledge Management; Vol. 12 Iss. 4, pp. 55-71.

83

Short Papers


A MEASUREMENT MODEL ON BLOGGERS’ SELF-DISCLOSURE Tang, Jih-Hsin Chen, Cheng-Chung Department of Information Management National Taipei College of Business Taiwan, R.O.C.

ABSTRACT The number of blog is growing explosively makes the question interesting: What do bloggers disclose themselves on the Internet? Drawn theories from social psychology, we develop a model to measure how deep and how wide issues that bloggers may express on their blog. Initial results showed that blog keepers might disclose nine topics such as general attitude, physical appearance, financial issues, career plans, emotion, self, interest, consumer experiences and others. However, only five of them are most frequently disclosed: career, emotion, consumer experiences, interest and others. Further studies may extend and validate the self-disclosure model (Altman and Haythorn, 1965) on the Internet. KEYWORDS Weblog, self-disclosure, measurement.

1. INTRODUCTION Blog is a Web-based personal publication system that allows users to add content easily and periodically, and normally in reverse chronological order on a personal website or community. Initially, blog is designed to express personal experiences, diaries, stories and knowledge exchange on the Internet(Cayzer, 2004; Nardi et al., 2004a; Nardi et al., 2004b); however, the popularity of blog makes all kinds of people keep their blog. It will become an increasingly important question of how to measure what a blogger’s disclose on his or her logs. While blog was coined in 1997 by John Barger, the number of blog is increasing explosively. The number of person who keeps blog, also called blogger, is increasing. The diversity of bloggers is also an interesting phenomenon. According to a report: the oldest blogger is 109, while the youngest blogger is 3 on earth. Why these people take their efforts in maintaining their blog remains an interesting question. Previous authors have different views on the classification of blogs. For example, Blood classified blog into three types: (1) filters: to record events on the world ; (2) personal journals: to maintain personal records; (3) personal opinions: to express opinions or emotion (Blood, 2004). Krishnamurthy proposed a typology of blog based on two dimensions: personal vs. topical and individual vs. community. To focus on the conversation nature of blog, we define blog as a personal online diary in this study. Drawn self-disclosure theories from social psychology, we develop a model to measure how deep and how wide issues that bloggers may express on their blog. We develop a measurement model and classify nine sub-scales of self-disclosure among bloggers. Initial survey results from 50 blog writers confirmed our model is valid and reliable. The contribution of this paper is that a measurement model for self-disclosure among bloggers is developed and validated, and further theoretical advance of how blog writers maintain their relationships among blog audiences, close friends and their parents may be developed.

87

ISBN: 978-972-8924-82-9 © 2009 IADIS

2. RESEARCH MODEL Self-disclosure is the sharing of intimate feelings and information with another person. Altman and Taylor proposed a model of relationship development, also called social penetration, is a central to self-disclosure (1973). Several authors have attempted to understand the factors of self-disclosure in the daily life(Jourard and Lasakow, 1958), the personal traits (Skoe, 1980), the conceptualization and measurement (Wheelees and Grotz, 1976), and on the Internet (Chiou and Wan, 2006; Fung, 2006; Qian and Scott, 2007) . Although the results are fruitful, they did not provide a detailed measurement model for bloggers. It seems valuable to design a self-reported scale for bloggers.

2.1 The Scale Development Based on previous literature(Fung, 2006; Jourard and Lasakow, 1958; Wheelees and Grotz, 1976; Yang and Hwang, 1980) and the top 50 listings of blog on Taiwan’s top blog service providers such as WRETCH.cc and Pixnet.org, we initially devised 108 items. We reworded and divided these self-reported items into 9 subscales: general attitude, physical appearance, career, emotion, self, interest, financial, consumer experiences and others. To measure the depth and breadth of each issue, we designed a 11-point Likert-like scale and the subjects were required to fill in the “depth” of how he or she may express on their blog. For example, a sample item was written as “I shall write down my point of views towards a news event”. The subjects were required to select 0%, 10%, 20% or 100% in order to show how deep they may express on this issue. If a subject choose 20% (willing to express 20% on the blog), then he shall get a converted score 8 on this item. While all subjects make their choices, compute the median score as intimacy value for each item. Each item get a intimacy value (Q3 – Q1). Only top 25% and lowest 25% items of intimacy values were chosen for refining the measurement.

2.2 The Subject Fifty subjects were recruited to pilot test the initial scale. All these subjects were blog writers. The demographics of the subjects were shown in Table 1. Twenty-two of the sample were male(44%), and twenty-eight female. On their educational level: 13 of them were junior and senior high school students, 23 on college and 14 were post-graduates. Table 1. Subjects’ Demographics Age 19 and under 20~24 25~29 30~34 35~39 40 and above Total

Frequency 9 15 12 7 3 4 50

Percentage 18% 30% 24% 14% 6% 8% 100%

3. CONCLUSION We validate the measurement model of self-disclosure on the Internet with reliability and validity analysis, and then present the most frequent disclosed topics, finally conducted preliminary analysis on some demographic variables such as gender and age.

88


3.1 Reliability and Validity As shown in Table 2, the Cronbach’ alpha is ranged from 0.62 to 0.95. And exploratory factor analysis demonstrated only one factor in each sub-scale. Therefore, the initial scale demonstrates fair validity and reliability. Table 2. Reliability of the scale Sub-scale Attitude Physical Financial Career Emotion Self Interest Consumer Experience Others

Cronbach α 0.90 0.91 0.94 0.95 0.62 0.95 0.95 0.94

Number of items 12 8 9 13 11 19 13 11

0.93

12

3.2 Most Frequent Disclosed Items Table 3. Most frequent disclosed Issues Sub-scale Career (work / school)

Emotion

Others Consumer Experience Interest

Items My holiday plans My extra-curricular activities or experiences at school My stress at work or at school What happened at school or at work today My gatherings with my classmates or colleagues How I get along with my classmates or colleagues My career or life plan My dreams (short-term or long-term) My feelings, emotion or feels Happy experiences or feelings My worries The moments in my down time or in low spirits My unpleasant or unhappy experiences My good memories My bad experiences My complaints My personal life Life trivial My traveling experiences (domestic or international) My gourmet experiences My favorite CD titles or my dislikes My opinions on movies or TV shows

Accumulated Frequency 37 74 35 70 35 70 35 70 34 68 31 62 30 60 30 60 42 84 42 84 39 78 37 74 37 74 35 70 35 68 34 68 36 72 30 60 31 62 30 60 32 64 32 64

On the depth and breadth of issues, we chose the most frequently disclosed items (with average accumulated percentage is above 60%). As shown in Table 3, 22 items were the most disclosed among blog writers. 8 items were on their work, school and career plans, another 8 items were on their feelings and emotions. Only two items were cited as the most frequently disclosed for consumer experiences and interests. Surprisingly, some sub-scales such as general attitudes, physical appearance, self and financial issues, no single item will be disclosed (accumulated percentage is above 60%) on their blog. On the gender and political issues, most bloggers were cautious to express their ideas. It seems that bloggers are aware of these issues to be private and are cautious to speak out in the “public area”.

89

ISBN: 978-972-8924-82-9 © 2009 IADIS

3.3 Gender Difference Although the small sample size, our data showed the gender difference on self-disclosure. For males, only 9 items are the most frequent disclosed; however, for females, 22 items are the frequently disclosed. For males, they tended to write down their ideas on the work, at school or career plans; however, the females tended to express their feelings, emotions and worries on the their blogs. Age seems to be a potential dominant factor for self-disclosure on the blog. For those who are age under 24, 31 items (of total 108) may be disclosed (with average accumulated percentage is above 60%); however, for those who are above 35, only 10 items may be expressed. It seems that older people might be cautious about their writings on the blog, yet the younger ones might be more straightforward.

3.4 Limitations We have developed a measurement model to gauge blogger’s self-disclosure on their blog, and initial results showed the model is fair. However, only pilot data have been tested. Additional data is to be collected. Furthermore, we plan to refine and shorten the 108 items into a brief version based on the intimacy score. The extended self-disclosure model may be developed and validated. The pilot study was conducted in one country and one culture that make the generability of the research findings limited. Further studies should be conducted in different cultures and countries and make the measurement of self-disclosure culture-free. Acknowledgements: the authors would like to thank National Science Foundation for their financial support (NSF )

REFERENCES Anonymous, http://worldsoldestblogger.blogspot.com/ Anonymous, http://worldsyoungestblogger.com/about/index.htm Altman, I., and W. Haythorn 1965. Interpersonal Exchange in Isolation. Sociometry Vol. 28, pp. 411-426. Altman, I., and D.A. Taylor. 1973. Social Penetration: the Development of Interpersonal Relationships Holt, Rinehart & Winston, New York. Blood, R. 2004. How Blogging Software Reshapes the Online Community. Communications of the ACM Vol. 47, No.12, pp. 53-55. Cayzer, S. 2004. Semantic Blogging and Decentralized Knowledgement Management. Communications of the ACM Vol. 47, No. 12, pp. 47-52. Chiou, W.-B., and C.-S. Wan. 2006. Sexual Self-Disclosure in Cyberspace among Taiwanese Adolescents: Gender Differences and the Interplay of Cyberspace and Real Life. CyberPsychology & Behavior Vol. 9, pp. 46-53. Fung, H.Y.P. 2006. Weblogging in Hong Kong: Motivations and Self-Disclosure. Master’s thesis, The Chinese University of Hong Kong, Hong Kong. Jourard, S., and P. Lasakow. 1958. Some Factors in Self-disclosre. Journal of Abnormal and Social Psychology Vol. 56, pp. 91-98. Nardi, B., D. Schiano, and M. Gumbrecht. 2004a. Blogging as Social Activity, or,Would You Let 900 Million People Read Your Diary? Communications of the ACM Vol. 6, No. 3, pp.222-231. Nardi, B., D. Schiano, M. Gumbrecht, and L. Swartz. 2004b. Why We Blog. Communications of the ACM Vol. 47, No.12, pp. 41-46. Qian, H., and C. Scott. 2007. Anonymity and Self-disclosure on Weblogs. Journal of Computer-Mediated Communication Vol. 12, pp. 1428-1451. http://jcmc.indiana.edu/vol12/issue4/qian.html Skoe, E.E. 1980. Personality Characteristics that relate to Self-disclosure: An Exploratory Study. Simon Fraser University:Master of Arts in the Department of Psychology. Wheelees, L., and J. Grotz. 1976. Conceptualization and Measurement of Reported Self-disclosure. Human Communication Research Vol. 2, pp. 338-346. Yang, M., and K. Hwang. 1980. The Vedge Model of Self-disclosure and Its Correlates. Acta Psy 22.

90


EXPERIENCING CHANGE THROUGH THE ADOPTION OF NEW DIGITAL TECHNOLOGY Mitul Shukla, Nik Bessis, Alfredo Gaitan Department of Computing and Information Systems, University of Bedfordshire, Luton, UK

ABSTRACT This paper discusses the experiences of change through the adoption of digital technologies into the work flow and business strategies from a sample Community of Practice. Various data collection techniques such as focus groups and interviews are identified and their implementation is described. The analyses take a Grounded Theory approach and are varied from content analysis (of the focus groups) to thematic/process analysis (interviews) and a more focused analysis of specially selected portions of material (axial coding). Finally, there is a thorough discussion of the findings of the data collected through multiple perspectives including organisational change, technology influenced change and the knowledge transfer process within changing organisations as evidenced by the changing workflow practices, business strategy and overall ‘shift in thinking’ adopted by these organisations. KEYWORDS Knowledge transfer, arts organisations, case study, digital technology.

1. INTRODUCTION The AmbITion project was a pilot programme intended to develop new competence in the use of Information and Communications Technology (ICT) within the arts/cultural sector. The project aimed to aid arts organisations to develop the effective use of digital technology across all areas of their business and artistic practice in an effort to improve their competitiveness and sustainability. AmbITion required the organisations to go through a process of technology based change, in project implementation as well as business strategy. The project helped to support organisational change through the implementation of new digital technology by 15 arts organisations situated in the North West and East of England (Ludus Dance, North West Disability Arts Forum, Royal Liverpool Philharmonic, Oldham Coliseum, New Writing Partnership, Firstsite, New Wolsey Theatre, Litfest, Lantern House, Cultural Fusion, Britten Sinfonia, Poetry Trust, Aldeburgh Productions, Colchester Arts Centre and Hoi Polloi). These organisations were allocated into two groupings: Tier 1 (7) and Tier 2 (8), based on the organisations level of commitment to the project. For example, Tier 1 organisations were expected to develop digital content as an integral part of their business model and implement a bespoke IT strategy. Tier 1 organisations would receive greater funding and assistance than Tier 2 organisations. It was decided that seven organisations would form a convenient sample for our study, namely for Tier 1, North West Disability Arts Forum, Oldham Coliseum and New Writing Partnership. With Tier 2 represented by, Litfest, Britten Sinfonia, Poetry Trust, and Hoi Polloi. The ACHIEVER (Assessing Cultural Handling of ICT; Enabling Value-Enhancing Responses) project is part of the larger AmbITion project and was implemented by the University of Bedfordshire. ACHIEVER was specifically tasked to assess the experience of the organisations involved with regards to the adoption of new technology. The main aim of this project then, was to gain an understanding of the experience of arts organisations when adopting new digital technology. Further aims included: documenting and theorising the organisational changes that takes place, as well as how knowledge transfer occurs. The emphasis was from the perspective of the organisations adopting digital technology, in contrast to a purely disciplinary approach. Within this context the aim for the rest of this discourse is twofold, firstly an examination of data collection techniques used and secondly a discussion of the findings made from the resultant data analysis.

91

ISBN: 978-972-8924-82-9 © 2009 IADIS

2. DATA COLLECTION The objective here was to assess the distance travelled by the organisations in terms of their expectations at the beginning of the project in October 2006 and the evaluation of outcomes as perceived by the organisations up to November 2008. A multi-method approach was adopted which allowed the researchers to collect data from a wide range of sources, from face-to-face interviews and focus groups to electronic means (blogs and e-mail) and documentation, see Bessis et al., 2008. This was combined with an ethnographic methodology which emphasised rigorous analysis of data and privileged the participants’ point of view. Although traditional ethnography includes prolonged and close interaction with a given community, the logistics of doing this with a disperse group, such as the sample, made this impractical. However, frequent contact with key participants from each organisation in the sample enabled the research team to maintain a friendly relationship with them and at the same time provided an effective means of obtaining information when necessary (via telephone or email). The first activity carried out comprised two focus groups, the first was for the East-based organisations (27th July 2007) and the second for the North West based organisations (9th October 2007). The aims were to: a) Create a baseline understanding of how the participants perceived themselves, their peers and their organisations in terms of the influence and use of ICT in their work and their experiences derived from organisational change. b) Expose the participants to a case study describing the use of ICT by an arts organisation and collect information regarding its perceived usefulness. Various elicitation techniques were employed in order to invite the participants to describe their experiences of change in the past, present and future. Participants followed a sequence in which they were encouraged to adopt three perspectives: individual, team and organisational. Participants were also provided with the opportunity to share their experiences in an informal peer group setting. A similar procedure was used with a focus on ICT. These were particularly useful in allowing participants to describe their experiences of moving across zones of peripheral participation in their organisations as communities of practice (Lave & Wenger, 1991; Wenger, 1998). Key members of the sample organisations described above were invited during December 2007 and January 2008 to participate in blogging about organisational change, learning, etc. Although they were encouraged through emails to write about specific themes, they also had the option to write about any other relevant topic related to, or stemming from, participation in the AmbITion project. Seven private blogs were set up using WordPress.com and were confidential to participants in the sense that only the researchers had access to them. The fact that the researchers also had a blog which the individual participants could see made blogging a reciprocal activity. The blogging was initially guided by a weekly email that proposed a topic for blogging. Finally, support was made available in the form of step-bystep guide on how to use the blog. It was hoped that participants would reflect about what was happening in their organisations as well as about partial findings from the focus groups fed back by the researchers. Site visits were also seen as a useful way to gain detail from the organisations. The purpose of the visits was to carry out interviews in which the participants could recount their experiences of AmbITion from their initial contact through to the various developments in the area of digital technology that had taken place at the time of the interview. Of particular importance were their expectations and hopes as well as their actual experiences of the AmbITion project. The interviews also offered an opportunity to engage in dialogue between the researcher and the participants on the topic of the process of knowledge transfer/generation and the amount of organisational change taking place or expected to.

2.1 Methodological Approach The analyses were based on Grounded Theory (Glasser & Strauss, 1967). They varied from content analysis (of the focus groups) to thematic/process analysis (interviews) and a more focussed analysis of specially selected portions of material (axial coding). Excel spreadsheets were extensively used in content analysis of focus groups where quotations/fragments of documents were recorded and coded. In this way they were easily accessed, searched and sorted by different codes depending on the stage of the analysis. Audio recordings were made of all the interviews with participants from the seven organisations and some of the consultants. These recordings were in turn transcribed in full.

92


Transcription and initial analysis of the interviews took place between June and August 2008. The thematic analysis that followed resulted in three core categories: Stage, Description and Evaluation (SDE), which allowed the researchers to make sense of statements from the interviews. Stages are broad time units in the participants’ experiences and attempt to capture the broad meaning of such experiences. They also help organise the most important or relevant details of events that occurred from the participants’ perspective (descriptions) as well as the impact of those events (evaluation). Once an SDE was produced it was shared by email with the interviewees for confirmation and clarification if required. Further feedback from the participants was sought, either by phone call, email or as a blog posting, and a number of questions that arose from the initial analysis were put to them. Questions were aimed at gaining detail about given situations within the organisation and were generally specific to each organisation.

2.2 Findings The findings presented here highlight the digital developments, knowledge transfer and the organisational change that have taken place. The work presented here is a summarisation of case studies produced for each of the sample organisations by the researchers. Additionally, some aspects of the experiences and processes are shared between organisations and it is possible to identify some trends.

2.2.1 Organisational and Technology Change Participants described the past situation in their organisations in the following terms: lots of unstructured information, difficulties (e.g. being inexperienced, lack of direction form management), adapting, growing and changing as well as the notion of audience development as an established feature. In relation to ICT, the participants mentioned difficulties with the development of an online presence (e.g. having a hard time in maintaining their website); workflow difficulties such as not having structured information storage, having fragmented IT systems or having a heavy reliance on e-mail. With regards to the present situation, participants described organisational change (e.g. rapid change, continuous new developments), sense of direction (e.g. new ideas becoming realised) and adaption to changes (e.g. centralising information). Of particular interest were ICT-related organisational changes taking place. These included: having a greater awareness of the need to profile audiences in order to engage with them; statements concerning workflow, for example, setting up procedures and documentation, further integration of staff and becoming more organised; and, online presence and skills exploitation typically to do with website growth or foraying into online social networking. The latter included knowing the level of inhouse skills available and what further training may be required in the case of skills exploitation. The participants envisioned their organisations experiencing a series of future changes. Their statements referred to three areas: firstly, specific aspects of the organisation such as increased internal communications; direction (e.g. working in the present, but looking towards the future) and growth (e.g. projects more aligned with wider goals). Secondly, the working environment, for example with regards to growing into a new building or the professionalism of colleagues. Thirdly, audience development, for example in terms of being further accessible to the audience and being able to inspire pride in people that live in the local area. Examples of how participants typically saw future ICT-related changes in the organisations included: more interactive and dynamic content and features on their websites and development of an online presence for greater audience participation. Participants also saw skills exploitation in terms of further learning and training as important. Emergent common themes from participant included: re-creating identity and role, formulating new marketing strategies and exploring the creation of digital content. The most ambitious aspect in terms of a vision was the desire to re-create identity and role as a construction of a ‘new presence’. However, the new presence also involved, for most organisations, ‘audience development’ aimed at improving relations with audience and for some, recording and transmitting festivals/events. Another means of recreating the organisation’s identity was described as becoming an intermediary between the artist and the audience. It was also recognised that for most of the organisations, their participation in AmbITion would have a direct impact on workload, roles, and workflow. Participants described an array of changes that occurred though their participation in the project, for example the staff at Oldham Coliseum could see that IT was being managed and developed in a planned way through the implementation of a ‘Technology Plan’. Although other organisations did not mention plans

93

ISBN: 978-972-8924-82-9 © 2009 IADIS

comparable to the one described by Oldham Coliseum, many of the components described by the ‘Technology Plan’ feature in other organisations accounts of change. For example, IT infrastructure renewal has taken place in many of the sample organisations.

2.2.2 Knowledge Transfer Knowledge transfer is seen here as referring to all exchanges of knowledge between individuals, teams and organisations. In the context of this project, IT-related knowledge is of special interest, but knowledge about artists, the arts organisations and audiences was also deemed as important. In some cases, the acquisition of knowledge became a goal in itself while in others, it made possible organisational changes that in turn enabled an organisation to relate better to its audience and to artists. Social network theory (e.g. Burt, 2004, 2007) is useful here as a point of reference, but our choice was to privilege the organisations’ perspective and the emphasis is on grounded theory that emerged from the analysis of the participants’ experiences in the AmbITion project. In trying to understand how knowledge transfer may have occurred an exploratory exercise limited to data from Britten Sinfonia was conducted. Participants described (a) ‘gaps’, i.e. perceived lack of activity in certain areas and therefore potential opportunities for knowledge transfer or application of knowledge, (b) ‘inhibitors’, i.e. characteristics that prevent knowledge transfer and learning, and (c) potential problems. However, it is useful to look at these at different levels. An example of gaps in knowledge, on the part of the organisation, was the insufficient understanding of the audience. In turn, the local audience, in the view of the participants, were not aware of international dimension of the organisation’s work, e.g. tours abroad or the performance of foreign artists and musicians. Some of the gaps in activity mentioned refer to the organisation’s website not having much interactivity, little or no rich content, e.g. blogs, videos or podcasts. Participants spoke about various conditions that were grouped under the term ‘inhibitors’ of knowledge transfer. Firstly, unclear demarcation of ICT responsibilities, means that people in the organisation did not know who to turn to when they required knowledge, wanted to relay knowledge or wanted to request or discuss some application of knowledge. Secondly, not tracking audience participation implied not having knowledge generated about the audience. The arts organisations approached by the AmbITion team started a process of planning which included identifying ‘training needs’ at personal, team and organisational levels. Individual needs refered, in the case of Britten Sinfonia, to IT and PR. As for the team there was an acknowledgement that although they had many aspirations, their practical knowledge was insufficient to produce the digital developments they envisioned. Britten Sinfonia, in their early documentation to AmbITion, had submitted a ‘Vision’ comprised of three components: (a) a ‘digispace’ which included podcasts, video on Youtube and a photographic archive; (b) a marketing dimension aimed at developing a relationship with their audience; and (c) an educational aspect (‘backing up the experience in the concert hall’ with content outside of concerts, such as, learning experiences, study days and masterclasses). An unintended and broader consequence of the planning and negotiation described above is the generation of ‘expectations’ about the possible and desirable outcomes of taking part in the AmbITion project and specifically around the exploration of the potential of digital technology. At the personal level participants from Britten Sinfonia expected to receive further training. They expected to see digital technology used in their normal workflow. They also hoped to see more learning resources and interaction on the website, as well as a new strategic plan for learning and raising funds. Some examples of knowledge gained were mentioned explicitly by members of Britten Sinfonia. In particular, individuals developed new skills related to the production of podcasts, and recording from external sources. Other individuals received training on a Content Management System (CMS) from a colleague. More general knowledge refers to lessons learnt from taking part in the AmbITion project.

3. CONCLUSION This paper discussed the experiences of change through the adoption of digital technologies into the work flow and business strategies from a sample of arts organisations. A review of data collection techniques to assess the level of change in 15 arts organisations over a 2 year period and appreciated by a sample of cultural organisations in adopting digital technology has been undertaken. Investigation was also conducted

94


into the realms of knowledge transfer and organisational change as impacted on organisations through the technology adoption identified. There has also been a discussion of the findings made from the data analysis as described. Consideration was made to gain an understanding of the experience of arts organisations when adopting new digital technologies. This was especially focussed on the distance travelled by the organisations in terms of their expectations at the beginning of the process and the eventual outcomes as perceived by the organisations up until November 2008. Finally, most organisations experienced a shift in thinking as a result of participating in the AmbITion project. For example, in the case of Litfest, the shift was from believing that their main role was in traditional publishing to considering themselves as multimedia publishers. From simply making documents available on their website in PDF format, they have started producing a wider range of materials available in various formats not onlyon their own website but also through media channels made available to them by others via the like of social networking sites such as Facebook, Flickr and Youtube. These materials can include information about authors and samples of work, as well as complete works that can be purchased online. Litfest has also had to take into account the ‘conservative’ nature of their traditional audience while also embracing the possibilities of a digital age.

REFERENCES Bessis, N., Gaitan, A., Shukla, M., Lai, Z. & Stephens, J. (2008). Using Next Generation Service Oriented Technologies to Assess Cultural Handling of Information communication technology, International Conference on Computer Science and Information Systems, Web Based Communities 2008, IADIS, Amsterdam, 24th -26th July. Burt, R.S. (2004). Structural holes and good ideas. American Journal of Sociology, 110, 2, Ppp. 349-99. Burt, R.S. (2007). Secondhand brockerage: evidence on the importance of local structure for managers, bankers, and analysts. Academy of Management Journal, 50, 1, pp. 119-148. Gaitan, A., Bessis, N., Shukla, M. and Stephens, J. (2009). Achiever: An Ethnographic Exploration of the Experiences of Change as Arts Organisations adopt New Digital Technologies, 5th International Conference on e-Social Science (sponsored by NCeSS), 24th-26th June, Cologne (under review). Glaser, B.G. & Strauss, A. L. (1967). The Discovery of Grounded Theory. Strategies for qualitative research. Chicago: Aldine. Lave, J. & Wenger, E. (1991). Situated learning: legitimate peripheral participation. Cambridge: Cambridge University Press. Pidgeon, N. (1996). Grounded theory: theoretical background. In J.T.E Richardson,. (ed.) Handbook of qualitative research methods. Leicester: B.P.S. Pidgeon, N. & Henwood, K. (1996). Grounded theory: practical implementation. In J.T.E Richardson, (ed.) Handbook of qualitative research methods. Leicester: B.P.S. Pidgeon, N. & Henwood, K. (1997). Using grounded theory in psychological research. In N. Hayes (ed.) Doing qualitative analysis in psychology. Hove: Psychology Press. Wenger, E. (1998). Communities of practice. Cambridge: Cambridge University Press.Our references go here (aim for around 5-10 of them)

95

ISBN: 978-972-8924-82-9 © 2009 IADIS

ANALYSIS OF ONLINE PARTICIPATION IN OPEN SOURCE COMMUNITIES M. R. Martínez-Torres E. U. Estudios Empresariales, University of Seville Avda. San Francisco Javier, s/n 41018 Sevilla (Spain)

S. L. Toral, F. Barrero E. S. Ingenieros, University of Seville * Avda. Camino de los Descubrimientos s/n, 41092, Seville, Spain

ABSTRACT The success of an Open Source Software (OSS) project is closely linked to the successful organization and development of the virtual community of support. In particular, participation is the most important mechanism in which the development of the project is supported. The main objective of this paper is to analyze the online participation in virtual communities using Social Network Analysis (SNA) techniques. Several Open Source communities related to Linux ports to embedded processors are studied, obtaining a set of indicators from their social network representation. The statistical analysis of these indicators allows the extraction of several conclusions about the online participation in this kind of virtual communities. KEYWORDS Open Source, Social Network analysis, virtual communities, factor analysis.

1. INTRODUCTION Open source software (OSS) is software whose source code is available to users and can be distributed with few limitations on possible modifications and distribution by third parties (Open Source initiative, 2002). In particular, OSS projects are developed and released under some sort of ‘open source’ license that allows inspection and reuse of the software’s source code (Crowston & Scozzi, 2002). Each OSS project is supported by a few, dozens or even hundreds of geographically distributed developers, organized as an Internet-based community, who voluntarily collaborate to develop the underlying software. The success of OSS projects has been attributed to their speed of development and the reliability, portability, and scalability of the resulting software. These claimed advantages of OSS development are due to the fact that the source code is open to the Internet community. Since everybody can access and review anybody else’s work, developers can learn from each other and improve their overall software development skill (Lussier, 2004). Also, instead of using huge financial resources to put the software through extensive testing and Quality Assurance, like a proprietary vendor will do, the Open Source projects have the community as a resource (Lakhani and Hippel, 2003). Finally, several studies claim that OSS is developed faster, cheaper, and the resulting systems are more reliable than proprietary software (Mockus et al., 2002). This paper will be focused on successful OSS development, but instead of analyzing the development in terms of the source code produced, we will focus on the social relationships among virtual community members. Virtual communities have become an important new organizational form and yet relatively little is known about the conditions which lead to their success. It is usually assumed that OSS communities are organized in a certain structure in which their members perform different roles according to their degree of involvement. This paper goes beyond this idea defining to what extent the structure and the different members’ profiles have an incidence on the successful development of the OSS community. Social network analysis (SNA) techniques will be used for this purpose. SNA models the community as a group of participants considering the links among them. From this model, several indicators measuring some features

96


of the community can be derived. A case study based on the Linux ports to different processors will be used to measure the proposed indicators and to identify the main dimensions involved in the development of the community. The obtained results could be applied to design or to improve OSS communities.

2. VIRTUAL COMMUNITIES IN OSS PROJECTS Virtual communities can be studied from the perspective of Communities of Practice (CoP) developed by Lave & Wenger (1991). This concept refers to the process of social learning that occurs when people who have a common interest in some subject or problem collaborate over an extended period to share ideas, find solutions, and build innovations. The basic assumption underlying the theory of CoPs is that engagement in social practice is the fundamental process by which we learn (Wenger, 1998). Learning in community should be viewed as an integral constituent of participation in the CoP, and as a process of constructing knowledge through social interaction with other members of the community, of changing relationships with other members of the community, and of transforming roles and establishing identities from a journeyman to a master in the community (Rohde et al., 2007). When interactions take place using electronic media, these communities are often referred as virtual communities or networks of practice (Johnson, 2001). OSS communities are typically initiated by an individual (or group of individuals) who provides systems and development components, or their access, as well as communication infrastructure. Participants are usually volunteers and contributors are not normally motivated by traditional economic incentives, but rather by instrumental factors associated with fulfilling a need, and by intrinsic factors such as enhanced reputation, expertise development (learning), self-fulfillment, as well as basic fun and enjoyment (Lerner and Tirole, 2002). The individuals that participate in open source software projects are often described as comprising a community. These communities have been described as having an onion-like structure, with a central core of highly active individuals, surrounded by other layers of progressively less active individuals. It has been demonstrated that much of the OSS development is realized by a small percentage of individuals despite the fact that there are tens of thousands of available developers. Such concentration is called “participation inequality” (Kuk, 2006), and it can be explained by the different user profiles of open source communities. Consequently, the structure of OSS communities is not completely flat as it was claimed by the bazaar model of full participation (Raymond, 1998). This argument is based on the logic that a highly participative community may lead to richer discussion, better flow of ideas, efficient code development, faster bug finding and fixing, and, hence, faster and efficient project growth (Weber, 2004). However, contrary to the bazaar view of OSS projects, many empirical studies have found that only a small number of developers contribute a large percentage of code and discussion in OSS projects. Participation inequality allows the categorization of OSS community members in three groups (Mockus et al., 2002): • Core members. They are responsible for guiding and coordinating the development of an OSS project. They are usually involved with the project during a long period of time and have made significant contributions to the development and evolution of the system. • Active developers. They regularly make contributions to the project. • Peripheral developers. They occasionally contribute with new features to the existing system. This contribution is irregular, and the period of involvement is short and sporadic. Free riders (people who just are seeking answers without making any contributions) are also included in this group. The roles and their associated influences in OSS communities can be realized only through contributions to the community. In this paper, we will analyze mailing lists because they allow the collective reflection and community discussions, and activities are not just confined to software development or coding alone (Sowe et al., 2006). Interactions are usually structured in threads of discussion, which facilitates their analysis. The simplest way to classify threads is using their length, i.e. the total number of posts they contain. Posts per thread turn out to be a reliable metric to determine the degree of “conversational concentration” of an author in a given group (Bonacci, 2004). Nevertheless, this kind of data does not provide any information about the social structure of the community, or about the relationships among authors. In this paper, social networks will be extracted from threads of discussion, and SNA techniques will be applied to provide new insights in the community organization (Cho et al., 2005). A social network can be represented as a graph G = (V,E) where V denotes a finite set of vertices and E denotes a finite set of edges such that E ⊆ V × V.

97

ISBN: 978-972-8924-82-9 © 2009 IADIS

In the context of threads of discussion, V is given by all the authors posting messages and E is given by the successive answers among authors inside a thread, which is the basic unit considered (Jones et al., 2004). The use of discussion threads as the basic unit of analysis is highly valid, considering that the epistemic interactions in support of OSS development often take place in discussion threads where individual postings provide the context to encourage participation (Kuk, 2006). In contrast to a reply to a single message, it is more cognitively complex to reply to a threaded discussion, because the ebb and flow of earlier postings must be taken into account to develop a coherent answer (Knock 2001). That is the reason why an author posting to a thread will be tied to all the authors who have previously posted to the same thread when constructing the social network. The resulting graph will exhibit the following features: • It will be a directed graph. The direction of the arc is given by the flow of information between two authors. That means that a sender (the tail of the arc) is answering a receiver (the head of the arc) inside a thread of discussion. • It will be a valued graph. An author is able to participate several times inside a thread or can answer to the same authors in different threads. Networks can be partitioned using some discrete characteristics of vertices. For instance, several classes of vertices can be obtained using the value of arcs. In the case of OSS projects, these kinds of partitions should highlight the core/periphery (C/P) structure of the community. A C/P structure divides vertices in two distinct subgroups: vertices in the core, densely connected with each other, and vertices on the periphery, not connected with each other, only nodes in the core. In network analysis, density is a measure of the cohesion of the network. More ties between people yield a tighter structure, which is, presumably, more cohesive. Density can be defined as the number of lines in a simple network, expressed as a proportion of the maximum possible number of lines. Consequently, maximum density is found in a network where all pairs of vertices are linked by two arcs, one in each direction. Sometimes, network density is not very useful because it depends on the size of the network. In this case, it is better to look at the number of ties in which each vertex is involved. This is called the degree of a vertex. As we are involved with a directed network, we will actually use the concept of out-degree of a vertex, that is, the number of arcs it sends. Therefore, the average out-degree of all vertices could be used to measure the structural cohesion of a network.

3. CASE STUDY The case study is based on Linux ports to embedded processors. Linux is a PC-based operating system that has been developed as Open Source Software along the structure of the UNIX operating system, and it is one of the most prominent examples of OSS projects. Although Linux started as a hobby in 1991, it represents today a serious threat to Microsoft Windows’s market dominance in operating systems (Cusumano and Selby, 1997). Nevertheless, the proposed case study will be focused on Linux ports to other processor architectures not intended for desktop or personal computer market. There are several reasons for this choice. First, Linux is firmly in first place as the operating system of choice for smart gadgets and embedded systems. Second, in contrast to other typical open source projects or even desktop Linux project, most contributions in this field do not come from volunteers or hobbyists, but from commercial firms, many of which are dedicated embedded Linux firms. Third, there are a lot of communities supporting each one of these Linux ports, and this is an excellent opportunity for analyzing a big group of more or less “homogeneous” communities. Up to eleven virtual communities have been considered: • The ARM Linux Project (ARM). ARM Linux is a port of the successful Linux Kernel to ARM processor based machines. • Debian port to ARM (D-ARM). ARM port for Debian GNU/Linux. Debian fully supports a port to littleendian ARM. • Linux PPC port (PPC). PowerPC Linux is the Linux kernel running on a PowerPC processor. • Debian port to PowerPC (D-PPC). PowerPC port of Debian GNU/Linux. The PowerPC architecture allows both 64-bit and 32-bit implementations. • Debian port to m68k (D-68k). Motorola 68k port of Debian GNU/Linux. Debian currently runs on the 68020, 68030, 68040 and 68060 processors. • Debian port to Alpha (D-Alpha). The purpose of this project is to assist developers and others interested with the ongoing project to port the Debian distribution of Linux to the Alpha family of processors.

98


• •

Debian port to MIPS (D-MIPS). MIPS port of Debian GNU/Linux, able to run at both endiannesses. Debian port to BSD (D-BSD). This is a port of the Debian operating system, complete with apt, dpkg, and GNU userland, to the NetBSD kernel. • Debian port to HPPA (D-HPPA). This is a port to Hewlett-Packard's PA-RISC architecture. • Debian port to Hurd (D-HURD). The GNU Hurd is a totally new operating system being put together by the GNU group. • Debian port to SPARC (D-SPARC). This port runs on the Sun SPARCstation series of workstations, as well as some of their successors in the sun4 architectures. Nine of them are Debian Linux ports to different processor architectures. The Debian Project is an association of individuals who have made common cause to create a free operating system called Debian GNU/Linux. The other two virtual communities are specific Linux ports to ARM and PowerPC processors. Each community will be analyzed during the period 2001-2007, which is the common period in which all the considered communities have been active. For each year and community, a social network based on interactions among participants has been extracted. As a result, a total of 77 social networks have been analyzed. The out-degree of each vertex will be used to distinguish among the different community members profiles. In particular, those members with an out-degree higher than the average out-degree of the social network will be considered as active contributors and those members with an out-degree higher than this average value plus the standard deviation will be considered as core members. Notice that these threshold values are chosen arbitrarily, but the important question for the subsequent analysis is to define a way of distinguishing the different members’ profiles independently of the size of the community. Using these general guidelines, the following variables can be extracted: • Community out-degree: out-degree of a social network represents the degree of interactions in threads of discussion. Consequently, average and standard deviation out-degree values (V1 and V2) will be obtained to be used as a threshold to distinguish among peripheral, active and core developers. • Active developers: the absolute value of active developers (V3) and their percentage respect to the whole community (V4) will be evaluated to consider the specific weight of this group. • Betweenness: it is a measure of centrality that rests on the idea that a person is more central if he or she is more important as an intermediary in the communication network (Nooy et al., 2005). The centrality of a person depends on the extent to which he or she is needed as a link in the chains of contacts that facilitate the spread of information within the network. The more a person is a go-between, the more central his or her position is in the network. If we consider that the shortest path between two vertices (geodesic) is the most likely channel for transporting information between actors, an actor who is situated on the geodesics between many pairs of vertices is very important to the flow of information within the network. The betweenness centrality of a vertex is the proportion of all geodesics between pairs of other vertices that include this vertex, and betweenness centralization of the network is the variation in the betweenness centrality of vertices divided by the maximum variation in betweenness centrality scores possible in a network of the same size (Nooy et al., 2005). Two values of betweeness will be considered: the betweeness of the whole network (V5) and the betweenness of the sub-network of active developers (V6). • Core developers: the absolute value of core developers (V7) and their percentage respect to sub-network of active developers (V8) and the whole community (V9) will be evaluated to consider the specific weight of this group. • Active and core developers out degree: the average out degree value of the sub networks of active developers (V10) and core developers (V11) are measures of participation inequality. The relative importance of the core will be measured evaluating the percentage of the out degree due to the core members of the community (V12), and their role as brokers (V13) or mediators among other core members. A factor analysis will be applied to extract the main dimensions related to online participation in virtual communities. Factor analysis has been performed using the principal component method. The eigenvalues of the sample covariance matrix are shown in Table 1. In factor analysis it is usual to consider a number of factors able to account for more than 70% of the total sample variance. In our case, study, this value is achieved with three factors.

99

ISBN: 978-972-8924-82-9 © 2009 IADIS

Table 1. Total variance explained

Factor 1 2 3 4 5 6 7 8 9 10 11 12 13

Eigenvalues 5,187 2,319 1,882 1,061 ,831 ,593 ,383 ,280 ,209 ,155 ,051 ,033 ,017

% of Variance 39,901 17,839 14,478 8,161 6,395 4,562 2,947 2,151 1,604 1,192 ,390 ,250 ,130

Cumulative % 39,901 57,740 72,218 80,379 86,774 91,336 94,283 96,434 98,038 99,230 99,620 99,870 100,000

Table 2. Rotated Component matrix with Varimax rotation.

V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11 V12 V13

F1 ,929 ,821 ,318 -,251 ,283 ,014 ,189 -,066 ,118 ,968 ,963 -,653 ,365

F2 ,226 ,143 ,699 ,563 ,196 -,244 ,923 -,174 -,135 ,104 -,009 -,138 ,866

F3 ,075 ,154 ,436 ,513 ,721 ,795 -,054 -,794 -,056 -,001 ,052 -,436 -,101

Using the associated eigenvectors, factor loadings can be estimated. Sometimes, it is difficult to perform the right interpretation of factors using the estimated loadings. Fortunately, factor loading can be rotated through the multiplication by an orthogonal matrix, preserving the essential properties of the loadings. Varimax method is an orthogonal rotation method that minimizes the number of variables that have high loadings on each factor. This method simplifies the interpretation of the factors. Table 2 reports the rotated factor loadings with varimax rotation for each one of the economical areas analyzed. To extract the meaning of each factor, we move horizontally through Table 2, from left to right, across the three estimated loadings of each variable, identifying the highest loading and the corresponding factor. To assess significance of factor loadings, a threshold value of 0,7 was considered. The association between variables and factors is highlighted in grey in Table 2. The resulting aggregation of variables leads to the following latent factors or dimensions: The first factor is explained by the participation inequality typical of virtual communities. This factor exhibits a high value in variable V1 and V2, which corresponds to the average and standard deviation of the out-degree of the network. The high value of the standard deviation means that there is a great variability in participation among community members. The high value of V10 and V11 confirm that the group of active and core developers are responsible of the majority of contributions. The second factor is related to the role of the core of the community. The core group is essential for the continuity of the community and they must play a brokerage role among contributors. The third factor is related to the topology of the community. Centrality (V5 and V6) and the structure of the communities (V8) are included in this factor. The negative value associated to V8 means that the core group should be just a small fraction of active developers, to guarantee a good coordination of the community. Several implications can be derived from the obtained latent factors: • The necessity of a participation inequality with a clear distinction between peripheral and active contributors. Open source communities are frequently visited by a lot of users just interested in asking for information, but with no intention of becoming an active contributor. Just a small fraction of visitors will become an active contributor as they learn through online participation. Learning does not appear as a result of being taught, but through direct engagement in the social, cultural, and technical practice of the community. • The key role of the core group. The mission of the core group is not just participating but, above all, promoting the debate and participation and addressing the future development of the underlying project. • Finally, a network structure is also necessary to achieve a good development of the community. The cohesion of the network is supporting the mechanism of participation, necessary for the project development, and participation promotes the success of the underlying project increasing the number of threads and contributions. On the other hand, cohesion also means a cohesive core group.

100


4. CONCLUSION Communities are basically based on interactions among users, and participation is the basic mechanism promoting its development. This participation has been analyzed using social network analysis techniques. Several indicators related to features like cohesion, structure, centralization and user profiles have been obtained for a set of online communities related to Linux ports and then analyzed using factor analysis. The obtained results reveal three main characteristics of open source virtual communities, like participation inequality, the role of the core group of developers and the necessity of a certain centralized structure around a small number of core developers.

ACKNOWLEDGEMENT This work has been supported by the Spanish Ministry of Education and Science (reference DPI2007-60128) and the Consejería de Innovación, Ciencia y Empresa (reference P07-TIC-02621)

REFERENCES Bonacci, D., 2004. Towards quantitative tools for analysing qualitative properties of virtual communities. Interdisciplinary Description of Complex Systems, Vol. 2, no. 2, pp. 126-135. Cho, H., Lee, J.-S., Stefanone, M. and Gay, G., 2005. Development of computer-supported collaborative social networks in a distributed learning community. Behaviour & Information Technology, Vol. 24, no. 6, pp. 435 – 447. Crowston, K. and Scozzi, B., 2002. Open source software projects as virtual organisations: competency rallying for software development. IEE Proceedings –Software, Vol. 149, no. 1, pp. 3-17. Cusumano, M. and Selby, R., 1997. How Microsoft Builds Software. Communications of the ACM, 40, no. 6, pp. 53-61. Johnson, C.M., 2001. A survey of current research on online communities of practice. Internet and Higher Education, Vol. 4, no. 1, pp. 45–60. Jones, G., Ravid, G. and Rafaela, S., 2001. Information overload and virtual public discourse boundaries. In Proceedings of Eighth IFIP Conference on Human-Computer Interaction, Tokyo, Japan. Knock, N., 2001. Compensatory adaptation to a lean medium: An action research investigation of electronic communication in process involvement groups. IEEE Trans. on Prof. Communication, Vol. 44, Iss. 4, pp. 267–285. Kuk, G., 2006. Strategic Interaction and Knowledge Sharing in the KDE Developer Mailing List. Management Science, Vol. 52, no. 7, pp. 1031–1042. Lakhani K. and Hippel E., 2003. How open source software works: ‘free’ user to user assistance. Research Policy, Vol. 32, no. 6, pp. 923–943. Lave J. and Wenger E., 1991. Situated learning: Legitimate peripheral participation. Cambridge University Press. Lerner, J. and Tirole, J., 2002. Some simple economics of open source. Journal of Ind. Economics, 50, 2, pp. 197-234. Lussier, S., 2004. New Tricks: How Open Souce Changed the Way My Team Works. IEEE Software, 21, 1, pp. 68-72. Mockus, A., Fielding, T. and Herbsleb, D., 2002. Two Case Studies of Open Source Software Development: Apache and Mozilla. ACM Trans. Software Eng. and Methodology, Vol. 11, no. 3, pp. 309-346. Nooy, W., Mrvar, A. and Batagelj, V., 2005. Exploratory Network Analysis with Pajek, Cambridge University Press, New York. Open Source Initiative, 1999, Open Source Definition, [online]. Source Available from: http://www.opensource.org/osd.html [last accessed Jan. 14, 2009, last updated July 24, 2006] Raymond, E., 1998. The Cathedral and the Bazaar, O’Reilly, Sebatopol, CA. Rohde, M., Klamma, R., Jarke, M. and Wulf, V., 2007. Reality is our laboratory: communities of practice in applied computer science. Behaviour & Information Technology, Vol. 26, no. 1, pp. 81 – 94. Sowe, S., Stamelos, I., and Angelis, L., 2006. Identifying knowledge brokers that yield software engineering knowledge in OSS projects. Information and Software Technology, Vol. 48, Iss. 11, pp. 1025-1033. Weber, S., 2004. The Success of Open Source, Harvard University Press. Wenger E., 1998. Communities of practice: learning, meaning, and identity, Cambridge: Cambridge University Press.

101

ISBN: 978-972-8924-82-9 © 2009 IADIS

SEARCHING FOR MARKET TRENDS AND BUSINESS OPPORTUNITIES IN WEB COMMUNITIES Graziella Martins Caputo, Nelson Francisco Favilla Ebecken Federal University of Rio de Janeiro Ilha do Fundão, CT, Bloco B, Rio de Janeiro – RJ, Brazil

ABSTRACT The objective of this paper is to examine if web communities can provide information about market trend and business opportunities given a set of services and product providers. Both Portuguese and Spanish languages were used to improve the results. Information extraction and web crawling were employed to find the most interesting subjects that interest a group of people. As case study, the web pages of two big enterprises that sell mobile products and services were analyzed. Their web pages were mapped, and then we evaluated what kind of costumer interest for their subjects and kind of others interests they have. With the result, it is expected to find relationships and new opportunities to increase customers’ interest and discover business opportunities. KEYWORDS Information Extraction, Web Mining, Market Trends, Business Opportunities.

1. INTRODUCTION To find information on the web is still a big challenge if you need to make a complex query. Fortunately, this hard work has been becoming easy by intelligent tools that can combine different terms, key-words, sentence and answer some specified questions. Despite of this evolution, is still difficult to find opinions about a product, a service or a brand. The same is applied to find business opportunities or market trends using simple queries that could follow what customers really need. Nowadays, large enterprises use the human knowledge and market research to find this information. Another source that is been used is the data available on the internet, where is possible to look for trends. These data can be found searching web communities (CHAU et al, 2005), where people are free to write what they think, need, or just what happens in their real life. A web community is defined by (Zhang, 2006) as a set of Web pages that has its own logical and semantic structures. In this way, the web page community considers each page as a whole object. In this paper, we focus on explicit-defined communities (Kumar, 1999), although it is possible to find implicitdefined communities to find information. The explicit web communities are groups of individuals who share a interests posting data on the web. They are useful because these communities can provide valuable, reliable, timely and up-to-date information source for a user. They represent the sociology of the web and are usually called Web 2.0 (Lee et al., 2008). It can be done by combining information about companies and people that interest for the same subjects. It can provide information that enterprises can use to improve their offer to better fit the costumer expectation or attract new customers. In this paper we suggest a methodology that, at first, map the main subjects of a market segment, and then, look for communities that have interest for the same subjects. We applied Information Retrieval in companies’ web sites, and explicit web communities’ pages. The relevance of the terms in the total set is extracted to represent the main subject that interests the customer.

102


2. METHODOLOGY The methodology consists of a set of steps that basically find the relevant information and filter all useless words and terms in the knowledge representation. For a given set of seeds, the web pages are crawled, the information is extracted, and then the important information is researched in the web. The starting point or seeds are companies’ web pages. The obtained information represents the business. It was used more than one enterprise aiming to diversify and does not focus on the same strategy. Both are of the same market segment, and compete for the same costumers. The first crawling step will download all the web pages in the domain of the seeds. Than an Information Retrieval (IR) tool extracts all relevant terms in the total set of documents downloaded. These terms are then searched in web communities using the second web crawling, aiming to find common interested people and what they use to associate with the subjects. All the relevant retrieved information is measured and it is estimated the value of each term inside the set of communities. These terms are the main content and can be considered as the interesting information to main subject.

2.1 Crawling Web crawling is used twice to find the web pages that satisfy the query criteria. First, the crawling tool is applied to store all the pages contained in companies’ web page. Second, the crawling tool is applied in web communities to find the related subjects of the main subjects of the companies web pages. For both companies’ web page and communities’ web pages, the algorithm stop criteria was the depth of links. When the depth achieves the level eight the crawling stops. For companies’ web pages, all the pages in the domain were downloaded and stored for the IR step. Links of external pages were not considered in the process, although it could have valued information. It will be considered in further works. Within all online communities it was used only that ones that could contain common interest in related subjects of the studied companies. For this, it was used blogs and personal comments in these blogs. The three web communities used as source of information are: 1) http://blogblogs.com.br/; 2) http://blog.uol.com.br/; 3) http://blog.terra.com.br/. The domain blogs were chose because they have a huge number of users. Each domain links for personal pages. Each personal page is normally about personal opinions and situations. These personal web pages, or blogs, are saved for further analysis and we call them document. These documents contain all the text written by common people, and also contain the comments that other people make about the written subjects.

2.2 Information Retrieval The IR (Moens, 2006) step is responsible to capture, in all documents, the terms and expressions that are significant to represent the subject. The IR step filters all the irrelevant data and just the relevant terms are considered to the next phase. The IR could help to find the needed information from the large collection of documents. As the crawling step, the information retrieval was applied twice in the process. The first in the companies’ web pages and second in the web communities web pages. The companies’ page IR phase aims to map the main subject business included in the company's strategy. For example, what their products, benefits, services, partners and others are. This is done extracting the main terms contained in the web sites. The tested sites are competitors. Both represent companies where their main business is mobile products and services. They have common interest and compete for the market share. They adopted different strategies and invest in innovation to their main business. So all the collected data was initially based on these two companies. After this IR process in the companies’ web pages, the communities’ web sites were crawled. The second IR step tried to find in the communities pages, the relevant subject that could improve the competitive intelligence of the companies, searching for trends and concurrency tips. The challenge of this phase was to chose the terms frequency inside each web community page. Some issues to find the perfect number of terms were made in this process:

103

ISBN: 978-972-8924-82-9 © 2009 IADIS

•

Documents (personal web pages) with low number of important terms, should not be dealing with the same theme. • It wasn’t necessary to find documents with all the selected terms, because they are rare and could restrict a lot the amount of documents. In this way, it was indexed documents that contain at least 4 of the relevant terms obtained in the first information extraction. At this context, all the relevant and useful information could be obtained and interpreted by a specialist.

3. EXPERIMENT RESULT The first crawling step selected 253 pages from both companies page. All the terms were indexed, and then applied text mining preprocessing techniques. This process include remove the stop-words and stemming applying (Markov, 2007). At the end, it lasted 4825 terms. Table 1 shows the most important terms. It includes proper names of products, services, partners, and The first column shows the common terms for both companies. The second column shows the terms that that just appeared in the first company web page and the third shows the second company unique terms. Table 1. Most relevant terms from the companies’ web pages Terms in Common 3G Activation and deactivation of services Bank Broadband Card Cell phone Chip Conference Control Download FAQ Games Gratis GSM Insurance International Roaming Internet Line Mobile National Roaming Online Online account Operator Plans Portability Prepaid Promotion SAC Signature Site Store Wi-fi

Terms of Vendor 1 Apparatus Audio and video Blackberry Bluetooth Brands and models Buy Car Change and recover password Contents Draw Email English Fiat Foston Gradient Help Interactivity Message News Offer Photos Reward Provedor Register Release Ringtones Spanish Synopsis Tips and tricks Uol Velocity Video maker

Terms of Vendor 2 Blocking call charge Broadband Call Identifier Call waiting Covenants Direct Debit Discounts Duplicate Environmental responsibility Equipment Fm Radio Franchise GPRS Installation Invoice Local call Lock loss and theft Migration Payment Consulting Personal care Phone PO Box Rates Recharge Signature Transfer Simulator Social responsibility Subdivision of account Suppliers Sustainability Technical Assistance Telephone for Deaf

All these terms can show the similar strategy of both enterprises and where they are offering different products and services.

104


The importance of each term is given by the TFxIDF measure, the most well known weighing scheme, where TF is the number of times that a term appears in a document and IDF is the inverse document frequency (LIU, 2006). From all the terms selected in the IR step, the 35 more relevant were chosen to the next step: find in web communities the personal pages that contain common subjects and interests. Table 2 shows the most relevant terms in the companies’ domain. Table 2. Main words from the total dataset Terms 3G Bill Account Service Pre-Paid Phone Question Roaming BroadBand Mobile Online Card

Relevance 0.599083 0.592268 0.582049 0.577217 0.563865 0.560787 0.492725 0.479002 0.478671 0.449064 0.432913 0.432913

Terms Plans Order Insurance Portability Regulation Virtual iPhone Promotion Video Track Recycle Highlights

Relevance 0.432061 0.419716 0.419716 0.414521 0.408235 0.403801 0.399389 0.398706 0.390852 0.383985 0.382108 0.377898

Terms Smartphones SAC Internet Post-Paid Message FAQ Internet Promotions Access Recharge Minutes

Relevance 0.374427 0.372364 0.370613 0.368209 0.364997 0.363042 0.358281 0.356061 0.353299 0.352416 0.350193

With the terms considered more important by the IR process, the domains were crawled looking for personal pages, or blogs, that have at least four of the terms listed in the table 2. From the three domain crawled, at least 550 personal pages contains four or more terms. All of the 550 pages were indexed and are now considerer a document, with the text written by the owner, and the foreign comments. Although all the domains of the blogs were Portuguese, some downloaded personal web pages were written in Spanish. The terms of these pages were indexed and faithful translated, without considering the language structure (Kawaba et al., 2008). Although it is a not recommended process, this work does not consider the structure of any page, so this method does not influence the results, on the contrary, just give more diversity for the costumer range. These words couldn’t be ignored because one of the companies has a huge market in Spanish language countries. This method increased the frequency of some word, and consequently, the relevance in the total set of documents. IR process reduced the number of terms to 2380, including the stop-words and stemming. Table 3 shows the 41 most important terms ordered by the relevance measured by the IR process. Table 3. Main words of the web communities Term Comment Music Service Entertainment Show Movie Photo Video News Blog Internet International Message Online

Relevance 0.033396 0.033165 0.032739 0.030021 0.02941 0.028939 0.028654 0.027994 0.027225 0.026472 0.025099 0.024512 0.023224 0.021792

Term Publish User File Web Page Theme Party Daily Electronic Guitar History Work Friendship Interview Event

Relevance 0.021668 0.021427 0.020277 0.019042 0.018506 0.017907 0.017731 0.017529 0.017484 0.017405 0.017277 0.017109 0.016677 0.01639

Term Festival Football Game Technology Issuing School Release Book Soup Opera News Prize Theater Television

Relevance 0.016387 0.016313 0.01604 0.015784 0.015752 0.015748 0.015695 0.015548 0.015455 0.015231 0.014911 0.014783 0.014766

We can notice that some final terms are the same of the initial, and some are different. The different terms shows the association that can be done with the main subject and related subjects. It means that the result

105

ISBN: 978-972-8924-82-9 © 2009 IADIS

shows the common interests between people and companies. It could be verified that using these terms the company can understand what interest their customer's have and then make improvement in products and services, or make new partnerships. It is also possible to make analysis to discover the actual boundary of the market and new trends. These results could help in the decision making about what kind of needs the costumers have. Some conclusions about the results could look obvious and, it means, that they correspond to a well known practice. But other conclusions could be analyzed to further improvement in the company offer. For example, to improve the mobile services with news and games about some terms like football, books, cinema and events with mobile and internet.

4. CONCLUSION Web communities are sites that people use to insert opinions and real life situations writing in blogs, e-mail lists, forums, and others web communities’ pages. It is possible to use this information to get opinions about products, services, brands and their needs. It is expected that part of the result information is obvious, but most of them are new and significant to improve the company strategy. It is also known that using more than one language can affect the results positively if the market demands for foreign countries. This methodology can be applied by any enterprise and e-commerce company that have a large number of interested and also based on the web. Combining web communities and information retrieval to find information for business application is a big improvement for the company competitive knowledge. It can be used as a tool for market trends, and possibilities to innovations if there is a necessity or advantage to the user. It can also, help to attract more customers, while satisfying their necessities.

ACKNOWLEDGEMENT We are grateful to the Brazilian Research Agencies CNPq and FAPERJ for their financial support.

REFERENCES Chau, M., Shiu, B., Chan, I., Chen, H., 2005b. Automated identification of web communities for business intelligence analysis. In: Proceedings of the Fourth Workshop on E-Business (WEB 2005), Las Vegas, USA, December. Moens Marie-Francine. 2006. Information Extraction: Algorithms and Prospects in a Retrieval Context, Springer. Zhang, Yanchun; et al.. 2006. Web Communities: Analysis and Construction. Springer. Markov, Zdravko; and Larose, Daniel T. 2007. Data Mining the Web: Uncovering Patterns in Web Content, Structure, and Usage. Wiley. Liu, Bing. 2006. Web Data Mining: Exploring Hyperlinks, Contents and Usage Data, Springer. Kawaba, M., H. Nakasaki, T. Utsuro, and T. Fukuhara. 2008. Cross-Lingual Blog Analysis based on Multilingual Blog Distillation from Multilingual Wikipedia Entries, ICWSM’08, Seattle, WA, March. Kumar, R., Raghavan, et al, 1999. Trawling the Web for emerging cyber-communities. Proc. Of the 8th International World Wide Web Conference. Lee, Sang-Heui, David DeWester and So Ra Park. 2008. Web 2.0 and opportunities for small businesses .Springer Berlin Heidelberg. Sclano F. and Velardi P. 2007. “TermExtractor: a Web Application to Learn the Common Terminology of Interest Groups and Research Communities “ 9th Conf. on Terminology and Artificial Intelligence TIA 2007, Sophia Antinopolis, October.

106


KNOWLEDGE SHARING THROUGH ONLINE COMMUNITIES OF PRACTICE: THE CASE OF LINUX PORTS TO EMBEDDED PROCESSORS S. L. Toral, F. Barrero E. S. Ingenieros, University of Seville * Avda. Camino de los Descubrimientos s/n, 41092, Seville, Spain

M. R. Martínez-Torres E. U. Estudios Empresariales, University of Seville Avda. San Francisco Javier, s/n 41018 Sevilla (Spain)

ABSTRACT The paper explores knowledge sharing processes in online communities analyzing their explicit and stored messages. Their content is processed using latent semantic analysis techniques, obtaining several indicators related to knowledge sharing activities. A factor analysis is then applied to obtain the main dimensions affecting knowledge sharing. The obtained results provide new insights in the underlying knowledge processes in online communities of practice. KEYWORDS Knowledge sharing, online communities, communities of practice, factor analysis.

1. INTRODUCTION The notion of community has been at the heart of the Internet since its inception (Lesser et al., 2000). Initially, Internet was used by scientist to share knowledge, collaborate on research and exchange messages, and today, million of Internet users worldwide communicate themselves using electronic tools. The advent of Web 2.0 and social software have propitiated the organization of users around communities of interest. The distinctive feature of these online communities is the intensive use of electronic media for people getting in contact. The theoretical background behind online communities has been treated by numerous authors. Some of them (Preece, 2001) highlight the connection of online communities with the social learning theory and communities of practice developed by Wenger (1998), while others are focused on their relation with knowledge sharing, knowledge creation, and innovation models (Lee and Cole, 2003; Kuk, 2006). Of particular importance are the online communities supporting OSS (Open Source Software) projects, as they are changing the way in which software is produced. Several case studies can be found on the literature. Mockus et al. (2002) raise some questions about OSS development and analyze two case examples based on Apache and Mozilla projects. Lee and Cole (2003) focused on the most well known OSS project: Linux. In this context, knowledge sharing means transforming individual knowledge into collective, organizational knowledge (Liebowitz, 2001). That is precisely one of the main objectives of a community of practice. Knowledge sharing is not stimulated by imposing structures and tools but by rich social interaction and its immersion in practice (van den Hooff & Huysman, in press). Consequently, knowledge sharing can be considered one of the most important processes involved in the development of an online community. The purpose of this paper consists of analyzing the main dimensions affecting knowledge sharing activities in online communities, using their stored information as a staring point. This information will be processed using latent semantic analysis techniques, extracting several indicators related to knowledge sharing, and these indicators will be statistically treated to identify the critical dimensions related to knowledge sharing.

107

ISBN: 978-972-8924-82-9 © 2009 IADIS

2. KNOWLEDGE SHARING In the early days of knowledge management, knowledge was often considered an object, which could be captured, codified and stored with the aid of information technologies. Knowledge management was mainly focused on optimizing these three processes. The problem of this approach is that the stored knowledge did not reflect real practices (Wenger et al., 2002). The insight that knowledge is not simply an aggregate of information that could be de-coupled from its context was then introduced and attention shifted to the idea that knowledge is socially embedded in the context where it takes shape (van Hooff & Huysman, in press). The online community can be defined as a social relationship aggregation, facilitated by Internet-based technology, in which users communicate and build personal relationships (Rheingold 1993). Individuals engage in knowledge sharing, problem solving, and learning through posting and responding to questions on professional advice, storytelling of personal experiences, and debate on issues relevant to the network. Examples of online communities can be found on fields like education, software development or consumer behaviour (Toral et al., 2005; Shang et al., 2006). Online communities have been frequently connected with communities of practice (Lin and Lee, 2006), in the sense that communities develop their own routines, formal and informal ‘‘rules’’, and practices evolve as a result of learning. The concept of Communities of Practice (CoP) was developed by Lave & Wenger (1991). This concept refers to the process of social learning that occurs when people who have a common interest in some subject or problem collaborate over an extended period to share ideas, find solutions, and build innovations. CoPs are not formal structures, such as departments or project teams. Instead, they are informal entities, which exist in the mind of their members, and are glued together by the connections the members have with each other, and by their specific shared problems or areas of interest (Wenger and Snyder, 2000). Intangible, tacit knowledge embedded in an organization's members is an asset that is not easy to capture. CoPs, however, offer a practical mechanism to help their members share and internalize tacit knowledge (Wang et al., 2008). Several researchers have noted that CoPs appear to be a more effective tool for dealing with unstructured problems and knowledge sharing/ creation than traditional and formal ways of structuring interaction in organizations. Understanding the processes and mechanisms that enable members to share knowledge in CoPs is very important for knowledge sharing within and between such communities (Pan and Leidner, 2003). A methodology based on semantic analysis technoques will be used for the identification of the main dimension with an influence in knowledge sharing. Modern semantic analysis techniques are based vector space model, in which documents are summarized and represented by vectors of words (term vectors). However, the main problem of such representation is the high dimensionality of the feature space (one dimension for each unique word). Therefore, it is desirable to first project the documents into a lower-dimensional subspace in which the semantic structure of the document space becomes clear (Cai et al., 2005). In the low-dimensional semantic space, the traditional clustering algorithms can be then applied. For instance, clustering using Latent Semantic Indexing (Zha et al., 2001) is one of the most well-known techniques. Different to LSI, in this paper we have applied the topic model, which is a statistical language model that relates words and documents through topics. It is based upon the idea that documents are mixtures of topics, where a topic is a probability distribution over words (Blei et al., 2003; Hofmann, 2001). Representing the content of words and documents with probabilistic topics has one distinct advantage over the purely spatial representation of LSI. Each topic is individually interpretable, providing a probability distribution over a word that picks out a coherent cluster of correlated terms. Hofmann (2001) introduced the probabilistic topic approach to document modelling in his Probabilistic Latent Semantic Indexing method (pLSI). Blei et al. (2003) extended this model by introducing a Dirichlet prior, calling the resulting generative model Latent Dirichlet Allocation (LDA). In both cases, the basic idea is that documents are represented as random mixtures over latent topics, where each topic is characterized by a distribution over words. Given T topics, the probability of the ith word in a given document can be written as T

P( wi ) =

∑ P(w |z

i i

= j ) P( z i = j )

(1)

j =1

where zi is a latent variable indicating the topic from which the ith word was drawn and P(wi|zi=j) is the probability of the word wi under the jth topic. P(zi=j) gives the probability of choosing a word from topics j in the current document, which will vary across different documents. Intuitively, P(w|z) indicates which

108


words are important to a topic, whereas P(z) is the prevalence of those topics within a document. Latent Dirichlet Allocation combines Eq. (1) with a prior probability distribution to provide a complete generative model for documents. The Latent Dirichlet Allocation algorithm involves the selection of several parameters to achieve a topic description of the domain under study. Table 2. Selected indicators Table 1. Dimensions involved in LDA algorithm. Parameter D W N L T ITER

Description Number of documents in corpus Number of words in vocabulary Total number of words in corpus Average length of document in words (L = N/D) Number of topics Number of iterations

Indicator Description I1 Number of topics I2 Polisemy I3 Rated Polisemy I4 Average messages size (characters) I5 Average number of messages per topic I6 Messages per topic distribution I7 Average messages size (words) I8 Size of threads I9 Average number of threads per topic I10 Threads per topic distribution

Table 1 details the key dimensions or size parameters describing a corpus (D, W, N and L) and a topic model run (T and ITER). In our study, we will analyze several online communities, extracting several indicators related to knowledge sharing by applying the topic model for each year. Selected indicators are detailed in Table 2. The first indicator I1 is the number of topics. This is a parameter of the topic model. Consequently, it has been chosen attending to the perplexity criterion. Perplexity is a standard measure of performance for statistical models of natural language (Manning & Schutze, 1999), and it is defined as: W ⎞ ⎛ 1 pplex = exp⎜ − log P( wn | d n ) ⎟ (2) n =1 ⎠ ⎝ W

∑

Perplexity varies from 1 to W; lower perplexity is better, and the maximum perplexity of W is reached when all words in the vocabulary are equally likely. In our case, LDA algorithm has been run for a number of topics varying between 1 and 50, selecting the number of topics leading to a minimum perplexity value. Indicators I2 and I3 are related to polysemy. Generative models have the ability of capturing polysemy, where the same word has multiple meanings. This is because they do not impose restrictions about mutual exclusivity that restricts words to be part of one topic. I2 measures the polysemy as the number of times a word wi appears more than once in different topics while I3 consider this previous value rated with the probability P(wi|zi=j) of the word wi under the jth topic. The next four indicators are related to messages. They consider the average size of messages in characters and words, the average number of messages per topic and their distribution over topics. Finally, the last three indicators are related to threads of discussion. It is usual that online communities are organized by threads of discussion. Threads are groups of messages sharing the same subject. A thread is initiated by someone who posts a message asking for help, suggesting some improvements, or just considering some new idea. Then people start answering this initial message, posting possible solutions, sources of information or just extending posted considerations. The indicator I8 relative to the size of threads considers those threads with al least one answer. Indicators I9 and I10 are the same than I5 and I6 but using threads instead of topics as the unit of analysis.

3. CASE STUDY The case study is based on Linux ports to embedded processors. Linux is a PC-based operating system that has been developed as Open Source Software along the structure of the UNIX operating system, and it is one of the most prominent examples of OSS projects. The same than Windows is the most prominent operating system released under a proprietary software license, Linux is the most prominent operating system released under a free license like GPL. Although Linux started as a hobby in 1991, it represents today a serious threat to Microsoft Windows’s market dominance in operating systems. Nevertheless, the proposed case study will be focused on Linux ports to other processor architectures not intended for desktop or personal computer market. There are several reasons for this choice. First, Linux is firmly in first place as the operating system

109

ISBN: 978-972-8924-82-9 © 2009 IADIS

of choice for smart gadgets and embedded systems. Second, in contrast to other typical open source projects or even desktop Linux project, most contributions in this field do not come from volunteers or hobbyists, but from commercial firms, many of which are dedicated embedded Linux firms. Third, there are a lot of communities supporting each one of these Linux ports, and this is an excellent opportunity for analyzing a big group of more or less “homogeneous” communities. Among the Linux distributions, Debian is perhaps one of the most well-known distributions. The Debian Project is an association of individuals who have made common cause to create a free operating system called Debian GNU/Linux, or simply Debian for short. Twelve Debian Linux ports communities have been considered for this study (Table 3). Each community will be analyzed during the period of time detailed in the description column of Table 3. Basically, the initial year is the one in which the community had a certain activity every month. For each year and community, indicators summarized in Table 2 were obtained. Consequently, 110 case studies were considered. Table 3. Virtual communities considered.

Community Debian port to m68k (D68k) Debian port to Alpha (DAlpha) Debian port to AMD64 (D-AMD64) Debian port to ARM (DARM) Debian port to BSD (DBSD)

Description Motorola 68k port of Debian GNU/Linux (98-08).

Community Debian port to Hurd (DHURD)

Description The GNU Hurd is a new operating system being put together by GNU group (99-08).

the Alpha family of processors port the Debian GNU/Linux (98-08).

Debian port to IA64 (D-IA64)

Debian port to Intel IA-64 (0108).

Debian port to MIPS (DMIPS)

MIPS port of Debian GNU/Linux, able to run at both endiannesses (99-08).

Debian port to PowerPC (DPPC)

PowerPC port of GNU/Linux (99-08).

Debian port to S390 (D-S390)

Debian port to IBM S/390 (01-08)

The port consists of a kernel for all AMD 64bit CPUs with AMD64 extension and all Intel CPUs with EM64T extension (04-08). ARM port for Debian GNU/Linux. Debian fully supports a port to little-endian ARM (99-08). This is a port of the Debian operating system, complete with apt, dpkg, and GNU userland, to the NetBSD kernel (01-08).

Debian

This port runs on the Sun SPARCstation series of This is a port to Hewlett-Packard's workstations, as well as some of PA-RISC architecture (01-08). their successors in the sun4 architectures (99-08). A factor analysis will be applied to extract the main dimensions related to knowledge sharing in virtual communities. Factor analysis attempts to identify underlying variables or factors, which can explain the pattern of correlations within a set of observed variables. Factor Analysis is a way to fit a model to multivariate data, estimating their interdependence. It addresses the problem of analyzing the structure of interrelationships among a number of variables by defining a set of common underlying dimensions, the factors, which are not directly observable, segmenting a sample into relatively homogeneous segments (Rencher, 2002). Factor analysis has been performed using the principal component method. The eigenvalues of the sample covariance matrix are shown in Table 4. In factor analysis it is usual to consider a number of factors equal to the number of eigenvalues higher than 1 or able to account for more than 70% of the total sample variance. In our case, study, this value is achieved with three factors. Using the associated eigenvectors, factor loadings can be estimated. Sometimes, it is difficult to perform the right interpretation of factors using the estimated loadings. Fortunately, factor loading can be rotated through the multiplication by an orthogonal matrix, preserving the essential properties of the original loadings. Varimax method is an orthogonal rotation method that minimizes the number of variables that have high loadings on each factor. This method simplifies the interpretation of the factors. Table 5 reports the rotated factor loadings with varimax rotation for each one of the economical areas analyzed. Debian port to HPPA (DHPPA)

110

Debian port to SPARC (DSPARC)


Table 4. Total variance explained. Eigenvalues 4,852

% of Variance 48,524

Cumulative % 48,524

2

2,821

28,214

3

1,861

4

Component 1

Table 5. Rotated Component matrix with Varimax rotation.

I1

F1 -,093

F2 ,959

F3 -,070

76,738

I2

,209

,951

,066

18,610

95,347

I3

-,069

,982

,033

,157

1,569

96,916

I4

-,117

-,016

,973

5

,099

,985

97,901

I5

,984

-,052

-,052

6

,081

,812

98,713

I6

,945

,066

-,011

7

,051

,513

99,227

I7

-,065

,036

,980

8

,036

,364

99,591

I8

,976

,120

-,078

9

,024

,236

99,828

I9

,977

-,081

-,069

10

,017

,172

100,000

I10

,962

-,008

-,124

To extract the meaning of each factor, we move horizontally through Table 5, from left to right, across the three estimated loadings of each variable, identifying the highest loading and the corresponding factor. To assess significance of factor loadings, a threshold value of 0,7 was considered (Rencher, 2002). The association between variables and factors is highlighted in grey in Table 5. The resulting aggregation of variables leads to the following latent factors or dimensions: • The first factor refers to topic activity. The activity around topics is highlighted by the high value of I5, I8 and I9 indicators which account the number of messages and threads associated to topics. However, the high values of the standard deviation in the messages and threads distributions per topics suggest that all the topics are not treated the same way. • The second factor is related to knowledge creation and reuse. The number of topics and polysemy are a measure of knowledge creation and reuse. This is precisely one of the main abilities of communities of practice. Topics are continuously evolving and previous knowledge is mixed and combined to generate new knowledge. • The third factor refers to the information provided. I4 and I7 indicators refer to the average size of messages. The availability and depth treatments of topics are also a determinant factor for a successful development of the underlying community. The obtained results demonstrate the necessity of guiding the evolution of the virtual community. Although virtual communities are based on the volunteer collaboration of community member, people posting a message hope to find an answer to their question or an alternative solution. Consequently, virtual communities are not only a question of social participation, but also a question of the quality of the provided information. Obviously, it is very difficult to asses individually the quality of each answer, as there are thousands of them, or to evaluate when a new knowledge is created or reused. For this reason, it is necessary to set a group of indirect indicators able to measure activity around the extracted topics, or the knowledge created through the evolution of the community.

4. CONCLUSION This paper analyzes knowledge sharing in virtual communities of practice. An automatic tool based on semantic analysis is proposed to avoid the implicit difficulties of analyzing individually thousands of messages to extract some measures about the community knowledge processes and activity. In particular, the Latent Dirichlet Allocation has been chosen as the specific algorithm to extract the topics in which the community is involved. As a case study, several communities related to Debian-Linux ports to embedded processors have been analyzed, measuring a set of predefined indicators. The application of a statistical technique like factor analysis allows the extraction of the main dimensions related to community knowledge sharing processes and activity. One of the main conclusions of this work is that community portal managers may use simple measures for evaluating actual virtual community stimulation.

111

ISBN: 978-972-8924-82-9 © 2009 IADIS

ACKNOWLEDGEMENT This work has been supported by the Spanish Ministry of Education and Science (reference DPI2007-60128) and the Consejería de Innovación, Ciencia y Empresa (reference P07-TIC-02621)

REFERENCES Blei, D. M., Ng, A. Y., and Jordan, M. I., 2003. Latent Dirichlet Allocation, Journal of Machine Learning Research, Vol. 3, pp. 993-1022. Cai, D., He, X., and Han, J., 2005. Document Clustering Using Locality Preserving Indexing, IEEE Transactions on Knowledge and Data Engineering, Vol. 17, no. 12, pp. 1624-1637. Hofmann, T., 2001. Unsupervised Learning by Probabilistic Latent Semantic Analysis, Machine Learning Journal, Vol. 42, no. 1, pp. 177-196. Kuk G. 2006, Strategic Interaction and Knowledge Sharing in the KDE Developer Mailing List, Management Science, Vol. 52, no. 7, pp.1031–1042. Lave J. and Wenger E., 1991. Situated learning: Legitimate peripheral participation. Cambridge University Press. Lee, G.K., Cole, R.E. 2003, From a Firm-Based to a Community-Based Model of Knowledge Creation: The Case of the Linux Kernel Development, Organization Science, Vol. 14, no. 6, pp. 633-649. Lesser, E. L., Fontaine, M. A., Slusher, J. A. 2000. Knowledge and Communities, Butterworth-Heinemann, pp. 86-98. Liebowitz, J. 2001. Knowledge management and its link to artificial intelligence, Expert Systems with Applications, Vol. 20, Iss. 1, 2001, pp. 1-6. Lin, H.-F., Lee, G.-G. 2006, Determinants of success for online communities: an empirical study, Behaviour & Information Technology, Vol. 25, No. 6, pp. 479 – 488. Manning, C. D., & Schutze, H. 1999. Foundations of Statistical Natural Language Processing, MIT Press, MA. Mockus A., Fielding R.T., Herbsleb J.D. (2002), “Two Case Studies of Open Source Software Development: Apache and Mozilla”, ACM Transactions on Software Engineering and Methodology, Vol. 11, no. 3, pp. 309-346. Pan, S. L. and Leidner, D. E. 2003. Bridging communities of practice with informaton technology in pursuit of global knowledge sharing. Journal of Strategic Information Systems, 12, pp. 71–88. Preece, J., 2001, Sociability and usability: twenty years of chatting online, Behaviour and Information Technology, Vol. 20, no. 5, pp. 347–356. Rencher, A.C., 2002. Methods of Multivariate Analysis. 2nd ed. Wiley Series in Probability and Statistics, J. Wiley & Sons. Rheingold, H., 1993. The Virtual Community: Homesteading on the Electronic Frontier, Reading, MA: Addison-Wesley. Shang, R.-A., Chen, Y.-C., and Liao, H.-J. 2006, The value of participation in virtual consumer communities on brand loyalty, Internet Research, Vol. 16, No. 4, pp. 398-418. Toral S.L., Barrero F., Martínez-Torres M.R., Gallardo S., Lillo J. 2005, Implementation of a Web-Based Educational Tool for Digital Signal Processing Teaching Using the Technological Acceptance Model, IEEE Transactions on Education, vol. 48, no. 4, pp. 632-641. van den Hooff, B., Huysman, M., in press, Managing knowledge sharing: Emergent and engineering approaches, Information and Management, in press. Wang, C.-Y., Yang, H.-Y., Chou, T., 2008, Using peer-to-peer technology for knowledge sharing in communities of practices, Decision Support Systems, 45, pp. 528–540. Wenger E. 1998, Communities of practice: learning, meaning, and identity, Cambridge: Cambridge University Press. Wenger, E., McDermott, R., and Snyder, W. M. 2002. Cultivating Communities of Practice. Harvard Business School Press, Boston, USA. Wenger, E.C. and Snyder, W.M., 2000, ‘Communities of practice: the organizational frontier, Harvard Business Review, Vol. 78 No. 1, pp. 139-44. Zha, H., Ding, C., Gu, M., He, X., and Simon, H. 2001. Spectral Relaxation for k-Means Clustering, Advances in Neural Information Processing Systems 14, pp. 1057-1064, Cambridge, Mass.: MIT Press.

112


CYCLIC ENTROPY OPTIMIZATION OF A SCALE-FREE SOCIAL NETWORK USING EVOLUTIONARY ALGORITHM Maytham Safar, Nosayba El-Sayed, Khaled Mahdi Computer Eng. Dept., Kuwait University

ABSTRACT We design and apply a Genetic Algorithm that maximizes the cyclic entropy of a social network model, hence optimizing its robustness to failures. Our social network model is a scale-free network created using Barabási and Albert's generative model. The maximum entropy equals 2.519 corresponding to an optimal network distribution found when the initial distribution is subject to 9 random edge removals and 11 additions of random edges regardless of the initial distribution. The most probable cycle size in the optimal solution is a cycle of 11 nodes. KEYWORDS Social Networks, Scale-Free Networks, Evolutionary Algorithm, Robustness Optimization

1. INTRODUCTION The current growth of information and communication technology has played an important role in changing the way people interact with each other. New forms of social interactions among people have emerged, including e-mail messages, bulletin boards and more recently, friendship networks (e.g. facebook, hi5, myspace, etc). Such networks could be described as “Social Networks” which consist mainly of groups of inter-connected people, where nodes represent individuals and edges represent the relationships between them. Social Networks analysis is mainly focused on studying the patterns of communication and exchange of information between people. These patterns are believed to have great influence not only on the individuals who adopt them, but also on the societies and organizations that enclose them. The spread of diseases, rumors, even beliefs, has proven to become strongly dependent on the inherent structure of the social network. An essential characteristic of any network is its resilience to failures or attacks, or what is known as its robustness. The definition of a robust network is rather debatable. One interpretation of a robust network is the network that assumes that social links connecting people together can experience dynamic changes, as is the case with many friendship networks such as Facebook, Hi5, etc. Individuals can easily delete a friend or add a new one, with and without constraints. Other networks, however, have rigid links that are not allowed to experience changes with time such as strong family networks. Entropy of a network is proven to be a quantitative measure of its robustness. Therefore, the maximization of a network’s entropy is equivalent to the optimization of its robustness. Albert et al. (Albert et al., 2000) describe the effect of a network’s heterogeneity on its degree of tolerance against either random node failures or intentional attacks. The three models of social networks are analyzed and compared: Scale-free (SF), Random Networks (RN) and Small-World (SW). Scale-free networks, which include social networks, were found to display a high degree of robustness against random failures but great vulnerability against targeted attacks. The study and analysis of resilience in complex networks was further investigated by many researchers, who mostly used percolation theory to study the resilience of different complex network topologies. Methods based on Percolation Theory focus on analyzing the threshold value pc, which represents the number of nodes that must be removed from a network before it disconnects into smaller, separate networks. Conversely, (Wang et al., 2005) studied the robustness of SF networks to random failures using entropy of the degree distribution in the network, hence the level of its heterogeneity. An optimal design of a robust network was achieved through the maximization of its entropy, following a nonlinear mixed integer programming approach.

113

ISBN: 978-972-8924-82-9 © 2009 IADIS

In this work, we maximize the entropy of a SF network using an evolutionary algorithm, hence finding the topology of the optimal SF network that is robust to random failures. We calculated the network's entropy as a function of cycles (cyclic-entropy). Hence, we find the optimal probability distribution function of the number of cycles that exist in the network and consequently the most probable cycle size. The originality of this work resides in the application of evolutionary algorithms to optimize networks' cyclic entropy.

2. ENTROPY OF NETWORKS Entropy is a very important characteristic that has been used to determine the degree of robustness in social networks (Mahdi et al., 2008, Safar et al., 2008, Sorkhoh et al., 2008, Wang et al., 2005). Entropy of a network is related to the probability of finding the network in a given state. For a system of moving molecules, the state is obviously the positions and the momentum of each molecule at a given instant. For a system of magnets, the state is defined through the magnets directed north or south. In social networks, there are several choices that define the state of the network; one is the number of social links associated with a social actor, known as degree. This definition is commonly used by almost all researchers. Characterization of social network through the degree leads to different non-universal forms of distribution. For instance, random networks have a Poisson distribution of the degree. Small-World networks have a generalized binomial distribution. SF networks have a power law distribution form. There is no universality class reported. Mahdi et al (Mahdi et al., 2008) proposes a universal distribution function form based the number of loops or cycles existing in the network instead of the degree of links in the network. The network configuration state was thus defined as the number of cycles within the network rather than the common definition of the network state as the degree of links associated with the actors in the social network. This new distribution form was found applicable to all types of social networks (scale-free, small world, and random networks). The same definition of the system state was used in (Safar et al., 2008) on a fully connected social network for the purpose of finding the maximum entropy value, hence identifying the equilibrium state of the social network, the state of maximum entropy. In other words, finding the point where the system is most stable. The classification of social networks was achieved by calculating different social network parameters (average vertex degree, clustering coefficient and average path) for a virtual friendship network, in (Sorkhoh et al., 2008). Loops were one of the major concerns in social network field. Loops (cycles) can be considered as the major aspect that can separate the graph to sub-graphs or components. A cyclic component is a group of intersecting cycles. They intersect with each other by lines or points. Other researches proved that there is a strong relation between the structural balance of a social network and the loops in the network. Balancing in social network is the collection of rules that defines the normal relations between the clients. Some networks must be acyclic. That is, it does not contain any cycles. Such a network is called a hierarchy. So, some researches were interested in detecting the invalid relation by counting the cycles in the network. This process gives each level in the hierarchy a number, a ranking process. From statistical mechanics, the entropy can be calculated from a given probability distribution P(k) of the system in state k: (1) S = − P (k ) ln P (k )

∑ k

where P(k) is the probability of finding a cycle of length k in the network. Based on the definition of entropy in Eq 1, we need to generate an initial distribution to start the optimization and let the evolutionary algorithm search for the optimal distribution that maximizes the total entropy of the network.

3. CYCLIC ENTROPY OPTIMIZATION Optimization algorithms are algorithms designed to find an optimal value of a variable x for a given function f(x), such that f(x) is either maximized or minimized, under a possible set of constraints. They are applied in solving optimization problems, which are mainly defined as problems of finding an optimal solution through searching within the set of all feasible solutions – also known as the solution space. Optimization problems are categorized as NP-hard (nondeterministic polynomial-time hard), where an exact solution for the problem

114


is difficult to find. Alternatively, heuristics are used to find a good solution that is close to the optimal solution of the problem, within a reasonable amount of time. Genetic Algorithms (GAs), one type of Optimization Algorithms, are adaptive heuristic search algorithms which are based on mimicking the powerful process of natural evolution. GAs are used in solving optimization problems to produce an optimal or near-optimal solution. The basic techniques of the GAs are designed such that the solution search process is directed into the region of better performance. This is accomplished by imitating the “survival of the fittest” principle, during the selection stage of the GA. Several generic selection algorithms exist, such as tournament selection, fitness proportionate selection and stochastic universal sampling. In this work, we use a different, evolutionary approach to optimize the entropy of the cyclic distribution in social networks: Genetic Algorithms (GA). We follow a comprehensive set of guidelines regarding the usage and design of a genetic algorithm for various problems, in (Baeck et al., 2000). Aspects such as fitness evaluation, constraint handling issues, population sizing and structuring, mutation parameters and parameter control are explained within the scope of conducted research in evolutionary algorithms. Moreover, the authors in (Baeck et al., 2000) discuss the efficient implementation of evolutionary algorithms for different sets of problems, using different parameters. In order to apply genetic algorithms effectively on SF networks, the structure of this type of networks should be examined first. Scale-free networks are networks that follow a power law degree distribution. In other words, the number of links “k” originating from a given node in a SF network exhibits a γ power law distribution: p(k) ≈ k , such that 2